3
Rise of the Private Sector
The central theme of this chapter is that most of both the current academic
literature and popular commentary on China’s economy have substantially
underestimated the contribution of the private sector to China’s economic
growth since 1978 while exaggerating the role of the state and state-owned
firms. As already noted, state-owned firms remain dominant in a number of
key sectors despite the far-reaching reforms of the 1990s. But, as this chapter
will show, overall private businesses grew very rapidly throughout the reform
era despite government policies that initially constrained their development.
The private sector is now the major driver of China’s economic growth,
employment, and exports and in recent years has even begun to contribute
to the increase in China’s outbound foreign direct investment. Moreover,
available evidence fails to support the frequently expressed view that China’s
2009–10 stimulus program, adopted in response to the global financial crisis,
directed credit disproportionately to state-owned companies, thus squeezing
out the private sector. To the contrary, on a number of metrics private businesses continued to outperform state-owned enterprises by a wide margin
both during and after the financial crisis, in part because of their improved
access to bank credit.
This chapter traces the rise of private companies and the decline of state
companies in agriculture, in industry and construction, and in services. After
reviewing changes in the ownership structure of the economy, the second
half of the chapter shows how the growth of private firms far surpassed the
growth of state companies after 1978, and it examines whether the party-state
increasingly controls the economy not through direct ownership of firms, as in
the past, but rather by recruiting private entrepreneurs to party membership,
thereby exercising control of the economy indirectly.
59
© Peterson Institute for International Economics | www.piie.com
Changing Structure of Ownership
Agriculture
Documenting the rise of private economic activity in the reform era is easiest
in agriculture. On the eve of reform, agricultural production was totally dominated by collective units called people’s communes. Commune leaders and
bureaucrats in government supply and marketing agencies in rural areas “were
deeply involved with all aspects of pre- and post-harvest decisions” (Huang,
Otsuka, and Rozelle 2008, 483). Outside of tiny private plots there was no
private use or control of agricultural land. Collectives were compelled to
deliver a large share of their output to state-owned procurement agencies at
prices fixed by the State Price Commission; the balance of agricultural output
produced on collective land was distributed in kind to commune members
based on individual labor contributions as measured through a system of work
points. Private local marketing activity and state-organized interprovincial
marketing of agricultural products were heavily restricted during the Cultural
Revolution (1966–76), reflecting both the general political bias against private
commercial activity and the official government policy of local food grain selfsufficiency (Lardy 1985, 43–45).
These statist arrangements had a profoundly adverse effect on the growth
of agricultural output, farm income, and levels of food consumption in
both rural and urban China. On the eve of reform in 1978 average per capita
consumption of grain and edible vegetable oils was slightly below and a third
below the respective levels of 1957, just prior to the Great Leap Forward when
farmers were organized into communes. Per capita rural income growth
between 1957 and 1978 was only 1 percent per annum in real terms (Lardy
1984, 850–52). Socialist agriculture led to stagnation in productivity and
incomes primarily because “neither the organization of production nor the
systems of pricing and marketing institutions provided suitable incentives”
(Huang, Otsuka, and Rozelle 2008, 477).
After 1978 the statist production and market arrangements of the prereform era were gradually transformed, giving rise to private farming. While
official policy at the outset of reform still called for collective farming arrangements, as early as 1978 Wan Li, who had become first party secretary of Anhui
Province in 1977, began to promote the decentralization of production tasks
to small groups or even to individual families. This decentralization, or as it
was called “contracting to households” (包产到户), quickly restored production incentives and greatly boosted agricultural production in the localities
in Anhui Province where it was adopted. In the fall of 1980 Wan Li, who by
then simultaneously held three powerful positions—vice premier, director of
the State Agricultural Commission, and the member of the Party Secretariat
in charge of agriculture—oversaw the drafting of an official government directive allowing contracting down to the household level throughout China.
By October 1981 over half of rural production teams, the lowest level in the
60
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
commune structure, had adopted some form of household contracting, and by
the end of 1982 this share rose to 98 percent (Vogel 2011, 436–37, 442).
Thus in the early 1980s collective production arrangements effectively
were disbanded throughout rural China, replaced by the household responsibility system in which farm land was assigned to individual families. Initially
land assignments were for relatively short periods, typically one to three years.
But to encourage private investment in land improvement, by the mid-1980s
the government extended the length of land leases to 15 years. Nonetheless,
these land use rights were somewhat insecure since local authorities could
reassign land within the contract period, discouraging investments to improve
land productivity.
The government subsequently instituted additional reforms to secure
farmers’ land use rights. These steps included extending the contract period
for land leases to 30 years in 1993, an initiative that was codified in the Land
Management Law in 1998. This was followed up with campaigns to issue
written land contracts. Government-directed efforts to improve land use
rights continued. In 2007 the Property Law declared that farmers’ 30-year land
use rights are private property rights. Subleasing of land use rights within
the remaining portion of the 30-year lease term is increasingly common, and
longer-term subleases have become more widespread in recent years. This
subleasing has allowed farmers to work outside of agriculture without losing
access to farm land. It has also made possible the emergence of private companies that are able to lease large amounts of contiguous land to operate largerscale agricultural production units (Prosterman et al. 2009).
Post-1978 changes in marketing arrangements were equally important. As
economic reform got under way, the state relied increasingly on price incentives rather than compulsory delivery quotas to acquire grain and other agricultural products required to feed China’s urban population. Free markets
began to reopen in large numbers in rural areas, and by the early 1980s free
markets were springing up in urban areas as well. The share of agricultural
output that farmers sold directly through these markets rose from less than
10 percent in 1978 to almost 20 percent by the mid-1980s (National Bureau
of Statistics of China 1983, 386; 1986, 472). Subsequently, further reforms,
including the abolition of rationing of grain and edible vegetable oil in urban
areas in 1993, led to a drastic reduction in the role of government parastatals
in marketing so that “today a large majority of sales of grains, oilseeds, and
fiber crops and virtually all of horticultural and livestock products flow to
small private traders,” a development that has led to increasing specialization
and productivity since the mid-1990s (State Council 1993; Huang, Otsuka,
and Rozelle 2008, 484).
The more market-oriented production arrangements adopted starting in
1978 led to a dramatic surge in agricultural production, food consumption,
and farm income. In the first five years of the new arrangements, “per capita
grain consumption rose 19 percent, vegetable oil consumption more than
doubled, and pork consumption rose 60 percent” (Lardy 1984, 857). Similarly,
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
61
between 1978 and 1985 rural per capita income growth soared by an average
of almost 15 percent per year in real terms, a huge multiple of the sluggish
growth between 1957 and 1978 (National Bureau of Statistics of China 1986,
555).
Only one small segment of agriculture was untouched by the transformations in agricultural production and marketing discussed above—state farms,
the origins of which date back to the 1950s. But this is a very small, rather
static universe. There were about 2,000 state farms in the mid-1960s, a number
that was basically unchanged through the early 2000s. By 2011 the number
of state farms had dropped slightly, to around 1,800. Similarly, the cultivated
area of these farms has changed little over the decades, but the number of
workers declined gradually from 5.1 million in 1978 to 3.3 million in 2011.
The main point is that state farms are a minuscule portion of Chinese agriculture, accounting in 2011 for about 1 percent of the agricultural labor force and
3 percent of the value of farm output (National Bureau of Statistics of China
1986, 173; 2001b, 395; 2012b, 125, 463, 495).
Although the ultimate ownership of rural land still lies in the hands of
village communities rather than individual farm families, farmers’ land use
rights have been improved dramatically since 1978. Given the transformation of production and marketing arrangements, agriculture (outside of state
farms) should be regarded as private and within the sphere of the marketized
portion of China’s economy.
Industry, Construction, and Services
Analyzing the transformation of the structure of ownership in China in industry, construction, and services is more complex than in agriculture for several reasons. Most importantly, these sectors cannot always be easily separated
into private and state domains. China has always had hybrid forms of ownership, including collectives, cooperatives, and joint ownership enterprises, that
do not always fall neatly into either the private or state category. And, as mentioned in chapter 2, in the 1990s the government introduced additional corporate forms of ownership such as limited liability companies and shareholding
limited companies.1 Without detailed information on the contributors of capital and the distribution of share ownership in these corporatized firms, it is
difficult to determine whether the majority or dominant owner of these new
corporate types is private, state, or some other type of legal person. Some early
analysis of reforms in industry was done based on a division of firms into two
broad categories, state and nonstate, where the nonstate sector includes most
or all of the firms with hybrid ownership arrangements (Jefferson and Rawski
1994). However, since state units were sometimes the underlying owner of
these hybrid firms, particularly collectives, this approach of using the nonstate
1. In some analyses the Chinese term for shareholding limited companies (股份有限公司) is translated as joint stock companies.
62
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
sector as a proxy for the private sector tends to overstate the transformation of
the ownership structure of Chinese firms. This is particularly true in the first
decade or two of economic reform, when hybrid forms of ownership were far
more important than in the past decade or so. Thus, with few exceptions, this
study avoids using nonstate as a substitute measure for private.
The second reason tracing the changing structure of ownership outside
of agriculture is difficult is that traditionally the National Bureau of Statistics
and other government agencies have organized and presented economic data
based on the legal registration status of individual firms. This creates problems,
as there are both state and private firms included within the most important
registration categories. For example, as noted in chapter 2, many traditional
state-owned enterprises have been transformed into limited liability companies or shareholding limited companies in which the state is the sole, majority,
or dominant owner. But within the categories of limited liability and shareholding limited companies, there are also firms in which the state has a very
limited or even no ownership stake whatsoever. Thus data on the output, assets, profits, and employment of limited liability companies and shareholding
limited companies cannot be attributed to state or private ownership without
more detailed information. Fortunately, it has become more common in recent
years for various Chinese bureaucracies to separate data into universes of firms
in which the majority or dominant shareholder is state, private, collective, or
foreign. This disaggregation allows us to peer inside the universe of firms registered as limited liability companies, shareholding limited companies, and
other registration categories to determine the underlying nature of the controlling owner or shareholder. The following discussion examines alternative
ways of measuring the size of the private sector in industry and services taking
into account these various lacunae in the definitions employed in official data.
The first and most frequent approach to measuring the size of the private
sector in industry and services in China is to rely on data that is based on the
official registration status of firms. All domestic firms in China must be registered.2 Industrial enterprises, for example, fall into one of seven broad categories, depending on their legal basis and ownership structure. Within most of
these categories there is a further breakdown of firm type depending on additional characteristics, such as the magnitude of the firm’s registered capital,
number of shareholders, and so forth. This scheme is summarized in table 3.1.
Registered private firms, one of the seven broad firm types, are further delineated into subcategories, shown in table 3.2. These include private sole proprietorship enterprises, private partnership enterprises, private limited liability
companies, and private shareholding limited companies. Firms of the first two
types are enterprises owned by a single individual or two or more persons in a
partnership, and in both cases the owners have unlimited liability for the debts
2. Even individual industrial and commercial households (个体工商户), referred to in this study
as “individual businesses” or “family businesses” and technically not considered to be firms
(企业), must register.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
63
64
MARKETS OVER MAO
Table 3.1
Enterprises classification system
Enterprise type
Asset ownership
Legal basis
State-owned enterprise (国有企业)
Owned by state
Law on Industrial Enterprises Owned by the
Whole People (1988); Law on the StateOwned Assets of Enterprises (2009)
Collective enterprise (集体企业)
Ownership shared by employees and other
economic entities
Provisional Regulations on Urban and
Township Collective Enterprises (1992)
Share cooperative enterprise (股份合作企业)
Shareholders are employees of enterprise;
initial startup using some public funds
Provisional Regulations on Urban and
Township Collective Enterprises (1992)
Joint enterprise (联营企业)
Jointly invested by two or more enterprise
legal persons or public institutions of the
same or different forms of ownership
Provisional Regulations on Urban and
Township Collective Enterprises (1992)
Limited liability corporation (有限责任公司)
Ownership based on capital contributions
Company Law (1994, revised 2006)
Shareholding limited corporation (股份有限公司)
Ownership based on shareholdings
Company Law (1994, revised 2006)
Private enterprise (私营企业)
Enterprise established by a natural person or
majority owned by a natural person
Provisional Regulations on Private Enterprises
(1988); Law on Wholly Individually Owned
Enterprises (2000); Law on Partnership
Enterprises (1997, revised 2007); Company Law
(1994, revised 2006)
Source: National Bureau of Statistics of China (2001a).
© Peterson Institute for International Economics | www.piie.com
Table 3.2
Private enterprise and household business classification system
Minimum capital
requirement
Tax status
Liability
Law
owned solely
by one natural
person
n.a.
individual
unlimited
Provisional Regulations on
Private Enterprises (1988),
Law on Wholly Individually
Owned Enterprises (2000)
Private partnership enterprise
(私营合伙企业)
shares are owned
by two or more
natural persons
n.a.
individual
unlimited
Provisional Regulations
on Private Enterprises
(1988), Law on Partnership
Enterprises (1997, revised
2006)
Private limited liability company
(私营有限责任公司)
ownership
based on capital
contributions
RMB30,000;
RMB100,000
(individual)
corporation
limited to
respective capital
contributions
Provisional Regulations on
Private Enterprises (1988),
Company Law (1994,
revised 2006)
Private shareholding limited company
(私营股份有限公司)
ownership based
on shareholdings
RMB5 million
corporation
limited to
respective share
holdings
Company Law (1994,
revised 2006)
owned solely
by one natural
person
n.a.
individual
unlimited
Enterprise type
Private sole proprietorship enterprise
(私营独资企业)
RISE OF THE PRIVATE SECTOR
Individual business
(个体户)
Asset ownership
n.a. = not applicable
Sources: State Council (2000; 2006b; 2011); National Bureau of Statistics of China (2001a).
65
© Peterson Institute for International Economics | www.piie.com
Provisions on Individual
Commercial Businesses
(1987, revised 2011)
of the company. The third and fourth types of private firms are limited liability
companies organized under the Chinese Company Law. In private limited liability companies the capital comes from a single person or jointly from two
or more persons. In private limited shareholding companies the capital is contributed by a single individual or jointly by 5 or more individuals.
The Chinese statistical authorities in the late 1990s started publishing
output and other data on registered private firms in the industrial sector and
for some components of the service sector. Thus there is information on the
number of such firms, their output, level of investment, magnitude of their
assets, and so forth. In some cases these data extend back to 1996. They show
that the number of registered private enterprises expanded from 443,000 in
1996, when they accounted for less than a fifth of all enterprises, to 5,918,000
in 2012, when they accounted for more than seven-tenths of all firms (National
Bureau of Statistics of China 1998; 2013c, 28–29). In the industrial sector the
number of registered private firms soared from 14,600 in 1999 to 273,000 by
2010, and the share of industrial output produced by these firms rose from 4.4
percent to 30.5 percent over the same period. Similarly, employment in registered private firms soared, exceeding 94 million by 2010, of which 33 million
were employed in industry (National Bureau of Statistics of China 2001b, 401;
2011a, 125, 499).
However, there are three major weaknesses in these official data on output
and employment based on firm registration status that result in a substantial
understatement of the importance of private firms in the industrial sector and
the economy more generally. Failure to take these weaknesses into account
has led a number of previous studies to underestimate the importance of the
private sector.
First, the National Bureau of Statistics began in 1988 to compile and annually publish data on the output of privately registered industrial firms only
for firms that have annual sales revenue above a fixed threshold, referred to as
“above-scale” (规模以上) firms. When China’s statistical authorities adopted
this practice, the cutoff point was RMB5 million; in 2011 they raised it to RMB20
million.3 As a result, the output of hundreds of thousands of small firms, which
are disproportionately private, is not included in this annually published data.
Fortunately, statistical authorities carried out two national economic censuses
that included these “below-scale” firms. The 2004 national economic census revealed, for example, that there were 947,000 registered private industrial firms,
almost eight times the 120,000 above-scale registered private industrial firms
(National Bureau of Statistics of China 2005b). While the National Bureau
of Statistics does not regularly report the output of these below-scale private
firms, the 2004 economic census did. Including the output of below-scale pri-
3. As a result of the increase in the threshold, the number of above-scale registered private industrial
firms fell between 2010 and 2011 from 273,259 to 180,612, and the number of employees in these
firms fell from 33.1 million to 29.6 million (National Bureau of Statistics of China 2012b, 501).
66
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
vate firms boosted the total value of industrial output of the private industrial
sector by three-fifths that year. Thus in that year above-scale registered private
firms accounted for 16.5 percent of industrial output produced by all abovescale firms, but expanding both the numerator and the denominator to include output of below-scale registered private firms and all below-scale firms,
respectively, boosts the private registered firms’ share of output to 22.4 percent
of total industrial output (National Bureau of Statistics of China 2005a, 488;
2006, 505).
The second reason annual data based on the official registration status of
firms fail to capture the extent of private economic activity in China, particularly in industry, is that many firms in which the majority or dominant owner
or shareholder is private (i.e., privately controlled firms) are registered in
categories other than private. This may arise for several reasons. As discussed
further below, in the 1980s some private firms registered as collective firms for
political reasons. In other cases private firms are registered in categories other
than private because their owners may wish to gain the advantage, for example,
of limited liability, which was not possible for solely owned private registered
firms until 2006. In other cases the ownership structure of firms evolves over
time, but some of these firms do not reregister after such a change has occurred
or do so only with a long lag (Dougherty, Herd, and He 2007, 312).
Another example of why the registration system fails to reflect the true
extent of private economic activity is that all firms in which a foreign company
contributes 25 percent or more of a firm’s capital are classified separately, with
a further subclassification based on whether the firm’s legal status is a joint
venture, a wholly foreign-owned firm, and so forth. This approach was adopted
to allow the government to track the success of China’s programs to attract
foreign investment, such as the law on joint ventures promulgated in 1979 and
the establishment of special economic zones on the southeast coast of China
beginning in 1980. While perhaps useful for this purpose, many of these firms,
particularly wholly foreign-owned firms in which the foreign owner is a private
corporation, in most economies would normally be regarded and classified as
private firms. This issue is further analyzed below.
An alternative approach to relying on a firm’s formal registration status
to track the changing structure of ownership in China’s economy is to look
at the ownership structure of firms and determine whether the majority or
dominant owner is private or state.4 This is possible because, as explained in
box 2.1, in 1998 the National Bureau of Statistics of China introduced the
concept of control and a year later began systematic publication of data on the
industrial output of firms that were registered in categories other than state
owned but were in fact majority or dominantly owned by the state. While the
government has not published such annual data on the output of privately
controlled industrial companies, the two economic censuses have provided
4. There are also a small number of limited liability companies and shareholding limited companies in which collective units are the majority or dominant owners.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
67
detailed ownership data for industrial firms that can be used to measure the
output of firms in which the majority or dominant share holder is private.
The earliest comprehensive published effort to measure the output of
privately controlled industrial firms was undertaken by economists at the
Organization for Economic Cooperation and Development (OECD) in collaboration with the National Bureau of Statistics of China. Using a comprehensive
firm-level data set covering a quarter-million industrial companies, they found
that privately registered and privately controlled domestic industrial firms
in 2003 accounted for a combined 33.1 percent of value added in industry, or
roughly 2.5 times the 12.8 percent private share reflected in official output data
based solely on firm registration status (Dougherty, Herd, and He 2007, 315;
National Bureau of Statistics of China 2005a, 488). The substantial understatement in the official data arises primarily because in 2003 privately controlled
firms accounted for about a third of the value added of firms that were registered
as collective enterprises and over half of the value added of firms registered as
other limited liability corporations.5 The OECD applied the same methodology
to a comparable data set for 1998 and found that privately registered and privately controlled industrial firms together accounted for 15.3 percent of value
added in industry, indicating that the share of output of this group of firms doubled in the five years leading up to 2003 (Dougherty, Herd, and He 2007, 315).
The same approach of classifying firms based on whether the majority
or dominant owner is private or state can be applied to firms with foreign
funding. The Chinese enterprise registration system does not classify firms
with foreign funding as private, state, or any of the other categories shown
in table 3.1.6 Rather, the registration system treats firms with foreign funding
amounting to 25 percent or more of the firm’s capital separately. In Chinese
statistics there are two categories of firms with foreign funding: those with
funding from Hong Kong, Taiwan, or Macao, and those with funding from
abroad.7 There are four variants within each broad category, depending on the
details of the ownership structure.
Should any of these firms with foreign funding be considered private? In
the early days of foreign-funded firms the answer to this question was basically
no. Through the mid-1980s virtually all foreign direct investment in China was
in the form of joint ventures. Although the domestic partner in some cases,
particularly for smaller joint ventures, was a private Chinese company, the
joint venture category was dominated by arrangements in which a state-owned
enterprise was the domestic partner of the foreign firm. Very frequently the
5. The line for limited liability corporations shown in table 3.1 is further divided into state
sole-funded corporations and other limited liability corporations. In 2004 the former category
included only 2,083 corporations, or 2 percent of the total (National Bureau of Statistics of China
2006, 505).
6. If the foreign ownership share is less than 25 percent, the firm is classified into one of the ownership categories shown in table 3.1.
7. In the text I sometimes refer to this combined universe of firms as “foreign funded.”
68
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
Chinese state firm was the majority shareholder in the joint venture. Thus for
the purposes of assessing the overall ownership structure within the Chinese
economy, firms with foreign funding through the mid-1980s should be considered predominantly state owned.
Over time this pattern changed. In 1986 the Chinese National People’s
Congress adopted a law governing foreign direct investment in the form of
wholly foreign-owned companies, in which there was no domestic ownership,
by either a state or a private firm.8 Subsequently, an increasing share of foreign
direct investment in China took the form of wholly foreign-owned firms. As
early as 1989, 30 percent of all incoming foreign direct investment was in this
form, up from only 1 percent in 1985 (National Bureau of Statistics of China
1990, 653). A decade later, the share had risen to two-fifths, and by 2010 just
over three-quarters of $106 billion in foreign direct investment entered China
in the form of wholly foreign-owned companies while less than one-quarter
was in the form of joint ventures (National Bureau of Statistics of China 2000,
605; 2011a, 244).9 In addition, in many cases foreign partners in firms that
were originally established as joint ventures with state firms subsequently
bought out their partners and converted their registration status to wholly
foreign-owned firms.10 The owners of these wholly foreign-owned firms are
private and thus should be regarded as private firms.
While foreign direct investment accounts for a very small share of capital
formation in China, in some sectors foreign-funded firms make a very large contribution to output. In the manufacturing sector, for example, foreign-funded
firms accounted for over one-quarter of the value of output in 2010, with about
15 percentage points of this produced by wholly foreign-owned companies.
Since manufacturing accounts for two-fifths of China’s GDP, the contribution
of wholly foreign-owned firms to the private sector should not be overlooked.
As early as 2003 two-thirds of the output of joint venture firms was produced
by firms in which the majority or dominant shareholder was private; only a
third of the output was produced by firms in which the majority or dominant
shareholder was a state or collective unit (Dougherty, Herd, and He 2007, 315).
A third major weakness of the official data is that they exclude a broad
range of economic activity carried out by what in China are referred to as
“family” or “individual businesses” (个体户).11 These businesses are not classi-
8. Law of the People’s Republic of China on Wholly Foreign-Funded Enterprises. The law was
revised in 2000, removing some restrictions on wholly foreign-funded firms, further encouraging
investment in that form of ownership. It should be noted that a few foreign investors established
wholly foreign-owned enterprises prior to the promulgation of the 1986 law.
9. The data cited in the text exclude foreign direct investment in China’s financial sector.
10. These ownership transformations are not reflected in the data just cited, which cover only new
investments.
11. The formal name for these businesses, “individual industrial and commercial households,”
clearly indicates that this category of economic activity excludes family-based agriculture. See note
2 in this chapter.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
69
Table 3.3
Individual businesses and registered private enterprises
in China, 2002–12 (millions)
Individual businesses
Year
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Businesses
23.8
23.5
23.5
24.6
26.0
27.4
29.2
32.0
34.5
37.6
40.6
Employees
47.4
43.0
45.9
49.0
51.6
55.0
57.8
66.3
70.1
79.5
86.3
Private enterprises
Enterprises
2.6
3.3
4.0
4.7
5.4
6.0
6.6
7.4
8.5
9.7
10.9
Employees
32.5
43.0
50.2
58.2
65.9
72.5
79.0
86.1
94.2
103.5
113.0
Sources: All-China Federation of Industry and Commerce (2011, 2; 2012, 30, 34; 2013, 26, 30).
fied as enterprises but are shown in the last line of table 3.2. Workers in these
businesses are classified as self-employed. Their numbers have grown enormously over time. In urban areas in 1978, the number of self-employed was
a minuscule 150,000, less than one-tenth of 1 percent of urban employment
(National Bureau of Statistics of China 1986, 92). This reflected the systematic
repression of private economic activity during the Cultural Revolution (1966–
76). By 2011 self-employed workers in urban areas numbered 52.3 million and
accounted for 15 percent of the urban workforce. In the same year there were
an additional 27.2 million individuals self-employed in nonfarming activities
in rural areas, up from 15 million in 1990, the first year for which data on
self-employed workers in rural areas have been published (National Bureau of
Statistics of China 2012b, 126–27; 1997, 97).12
While official data on registered private firms exclude the output and
employment of individual businesses, other data sources provide a more
comprehensive picture of private firms and individual businesses, including
those outside the industrial sector. Table 3.3 shows the expansion since 2002.
By 2012 these businesses employed 200 million, two and a half times the
level of 2002. The sectoral breakdown in 2009 is shown in tables 3.4 and 3.5,
12. However, the rise from 1990 to 2010 was not linear. Reported self-employment in rural areas
peaked at 39 million in 1998. The decline from this peak presumably reflects several factors. First,
the rapid development of the export processing and urban construction sectors led to a large
increase in the number of rural residents who migrated to urban areas to seek employment. While
some of these migrant workers had previously been employed in agriculture, others presumably
had been employed in township and village enterprises. Second, some household businesses in
rural areas presumably converted to registered private enterprises once they reached the limit of
seven nonfamily employees.
70
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
respectively. The most striking thing to emerge from both tables is that private
firms and individual businesses are concentrated in the service sector and that,
within services, wholesale and retail activity dominates, as measured by the
share of firms and share of employees in private and individual businesses.
Manufacturing accounts for almost a quarter of private firms, but fewer than
10 percent of all employees in individual businesses. Given the restriction that
individual businesses until 2011 could not have more than seven nonfamily
employees, it is not surprising that their role in manufacturing, where economies of scale are usually significant, is so limited.
Figure 3.1, which summarizes the discussion above, shows the complexity
of answering the seemingly simple question “How large was the private industrial sector in China in 2003 and 2007?” The answer depends on how one defines private. On the narrowest definition, based solely on firm registration
status, above-scale registered private firms accounted for only one-eighth of
China’s industrial output in 2003. But taking into account the output of both
private registered firms with revenue less than RMB5 million and domestic
firms registered in categories other than private but in which the majority or
dominant shareholder was private brings the private share to almost two-fifths
of all industrial production. Adding in the output of firms with foreign investment, where the majority or dominant shareholder is private, pushes the total
up to almost three-fifths. The data for 2007 show a similar pattern, but the
private share is somewhat higher for each definition of private, reaching over
three-fifths on the broadest definition. In both years the narrowest definition
clearly understates the contribution of the private sector to industrial output,
but the broadest probably overstates it because there must be some firms where
the extent of the government’s de facto control exceeds its ownership share.
In the following analysis of the importance of the private sector in generating output, employment, and exports, the definition of private is the sum of
registered private firms and individual businesses. It would be useful to be able
to base the analysis on the concept of privately controlled firms, following the
lead of the OECD. However, the large firm-level data set required to measure
the extent of private control of firms with various registration statuses is
available only for years in which China’s statistical authorities conduct an
economic census, typically every five years, and even these data become available with a substantial lag.13 Moreover, the OECD has published its findings
for only two years, 1998 and 2003, and thus cannot shed light on developments in the decade since. Finally, the large firm-level data set that the OECD
used to determine control only covered firms in the industrial sector, and not
ownership transformation in services. As a result, this analysis initially relies
on a narrower definition of private, but the reader should keep in mind that
13. Summary data from the 2003 census and the 2008 census were released at the end of December
2005 and the end of December 2009, respectively. The more detailed microdata from each
economic census were available with an even longer time lag.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
71
Table 3.4
Private enterprises by sector, 2009 (thousands)
Total
Primary
Farming, forestry, animal husbandry
Secondary
Manufacturing
Mining
Construction
Electricity, heat production, and supply
Tertiary
Wholesale and retail trade
Leasing and commercial services
Scientific research and polytechnical services
Information transmission, software, and information technology service
Real estate
Resident services and other services
Transport, storage, and postal services
Accommodation and catering trade
Culture, sports, and entertainment
Other
Water conservancy, environment, and utility management
Finance
Health, social welfare, and social services
Education
7,402
164
164
2,206
1,745
60
367
35
5,032
2,632
717
379
274
249
243
195
137
86
48
30
24
10
9
Source: All-China Federation of Industry and Commerce and China Nonstate (Private) Economy
Research Association (2011, 738).
the result understates private businesses’ role in China. (Later in the chapter,
in the section explaining the growth of the private sector, I do examine official data series on investment, lending, and other variables that are based on a
broader definition of private.)
Industry and Construction
The rise of private firms and the decline in the role of state firms is easiest to
measure in industry and construction. Until 2013 industry and construction
combined accounted for a larger share of China’s GDP than either agriculture or services, so understanding the changing ownership arrangements in
industry and construction is crucial to any evaluation of the evolving relative
roles of state and private firms in the Chinese economy. As shown in figure 3.2,
the share of industrial output produced by state firms (including both traditional state-owned enterprises and enterprises in other registration categories
where the state is the majority or dominant shareholder) fell from 78 percent
in 1978 to 26 percent by 2011 (National Bureau of Statistics of China 1990,
416; 2012b, 501, 513).
72
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
Table 3.5
Employees of individual businesses by sector, 2009
(thousands)
Total
Primary
Farming, forestry, animal husbandry
Secondary
Manufacturing
Mining
Construction
Electricity, heat production, and supply
Tertiary
Wholesale and retail trade
Resident services and other services
Accommodation and catering trade
Transport, storage, and postal services
Leasing and commercial services
Other
Information transmission, software, and information technology service
Culture, sports, and entertainment
Health, social welfare, and social services
Real estate
Scientific research and polytechnical services
Education
Water conservancy, environment, and utility management
Finance
31,974
427
427
2,776
2,640
53
67
16
28,801
19,903
3,245
2,813
1,596
374
289
210
196
87
45
26
9
7
1
Source: All-China Federation of Industry and Commerce and China Nonstate (Private) Economy
Research Association (2011, 740).
But this series understates the decline of state industry because, as already
noted, data on the value of industrial output are available on an annual basis
after 1998 only for above-scale firms. The abrupt jump in state-owned firms’
share of industrial output in 1998, reflected in figure 3.2, is entirely a statistical
artifact driven by the deletion of below-scale firms, which were overwhelmingly
private firms or collective enterprises.14 For 2004, China’s statistical authorities published data including below-scale firms. Thus, while the figure shows
that state firms in 2004 accounted for 35 percent of above-scale firms’ industrial output, taking into account the additional RMB3.5 trillion in output
produced by below-scale firms drops the state share to 30 percent. A little over
half of the output of below-scale firms was produced in private enterprises.
14. From 1998 through 2006 the output value of above-scale industrial firms included the output
value of all state firms and the output value of nonstate firms with sales revenue of RMB5 million
or more. For the four years 2007–10, it includes the output value of all firms with sales revenue
of RMB5 million or more. Beginning in 2011 it includes the output value of all firms with sales
revenue of RMB20 million or more.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
73
74
MARKETS OVER MAO
Figure 3.1
Alternative measures of the private share of value-added industrial output, 2003 and 2007
62.7
Privately controlled, including firms with foreign capital and below-scale firms
55.9
59.7
Privately controlled, including firms with foreign capital
52.3
41.2
Privately controlled, including below-scale firms
36.7
38.2
Privately controlled above-scale firms
33.1
25.5
Privately registered firms, including below-scale
16.4
22.5
Privately registered above-scale firms
2007
2003
12.8
0
10
20
30
40
50
60
70
percent share of value-added industrial output
Sources: Author’s calculations based on data from the National Bureau of Statistics of China, Industrial Survey Firm-Level Database (2007); Dougherty, Herd, and He (2007, 315).
© Peterson Institute for International Economics | www.piie.com
Figure 3.2
Gross industrial output of state enterprises, 1978–2011
percent of total gross industrial output
90
80
70
60
50
40
RISE OF THE PRIVATE SECTOR
30
26
20
10
0
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
75
Notes: Data include only above-scale firms, with sales greater than RMB5 million (1998–2011) and firms with sales greater than RMB20 million (beginning in 2011). This data series appears
to have been discontinued.
Sources: National Bureau of Statistics of China (2012b, 518); National Bureau of Statistics of China, www.stats.gov.cn (accessed on February 28, 2014).
© Peterson Institute for International Economics | www.piie.com
Thus in 2004 private firms’ share of all industrial output was 22.4 percent, well
above the 16.5 percent share of above-scale private firms’ output. In short, the
official data underlying figure 3.2 accurately reflect the long-term decline in
the role of state firms in China’s industrial sector, but after 1998 their share
of output is overstated. If the relative importance of below-scale firms in 2011
was the same as in 2008, state firms’ share would be 24 percent, rather than the
26 percent shown in the figure.15 Roughly speaking, it is therefore fair to say
that between 1978 and 2011 the share of industrial output produced by state
firms fell from about three-quarters to one-quarter.
The retreat of state firms in Chinese industry has been across almost all
product lines. By 2011 there were only 6 of 40 branches of industry where
state firms produced more than 50 percent of output (table 3.6). Two of the
six, electric power and water, are natural monopolies in which state ownership is common in many market economies. In OECD countries, for example,
state-owned utilities account for almost 30 percent of the value of all stateowned companies (Christiansen 2011, 15). Moreover, in 24 branches of industry the state share of output was less than 15 percent. Even these data
overstate the role of state firms since the output data exclude firms with sales
of less than RMB20 million, a universe that comprises predominantly private firms.16 Especially worth noting is the significant erosion of state-controlled firms’ role even in some of the nine key “pillar” industries discussed
in chapter 2. In the coal industry, the state’s share of output by 2011 had
slipped to barely over 50 percent, down significantly from the 66 percent share
it held in 2006, when Li Rongrong declared that it was necessary for the state
to guarantee absolute controlling power in this industry. In steel, another socalled pillar industry deemed as requiring strong state control, the share of
output by state-controlled firms slid from 43 percent in 2006 to 37 percent in
2011 (National Bureau of Statistics of China 2007, 501, 512; 2012b, 502–503,
512–13).
While China’s three national oil companies continue to control over ninetenths of oil and gas extraction, the one-time dominance of these firms in
some downstream activities has been significantly eroded. The best example
is natural gas, where state firms by 2011 controlled less than half of gas distribution. As early as 2002 the distribution of natural gas in China’s cities was
opened to private players, including foreign companies. There are now more
than 60 companies in the city gas business. Some are state-owned, including
China’s largest gas company, Beijing Enterprise Holdings Limited. Others are
15. Data for the gross value of total industrial output and the gross value of industrial output of
above-scale firms are available for 2008 since statistical authorities conducted an industrial census
that year.
16. When the threshold for above-scale changed in 2011, the number of state above-scale firms fell
by 3,200 between 2010 and 2011, while the number of private above-scale firms dropped by about
90,000 (see note 14). These numbers suggest that the data in table 3.6 overstate the role of state
firms in industry.
76
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
Table 3.6
State share of industrial output by subsectors, 2011
Sector
Manufacture of tobacco
Production and supply of electric power and heat power
Extraction of petroleum and natural gas
Production and supply of water
Processing of petroleum, coking, processing of nuclear fuel
Mining and washing of coal
Production and supply of gas
Manufacture of transport equipment
Smelting and pressing of ferrous metals
Smelting and pressing of nonferrous metals
Mining and processing of nonferrous metal ores
National average
Manufacture of special purpose machinery
Manufacture of raw chemical materials and chemical products
Mining and processing of ferrous metal ores
Manufacture of beverages
Manufacture of general purpose machinery
Mining and processing of nonmetal ores
Manufacture of rubber
Manufacture of medicines
Printing, reproduction of recording media
Manufacture of nonmetallic mineral products
Manufacture of measuring instruments and machinery for cultural
activity and office work
Manufacture of electrical machinery and equipment
Manufacture of artwork and other manufacturing
Manufacture of communication equipment, computers, and other
electronic equipment
Manufacture of chemical fibers
Manufacture of paper and paper products
Manufacture of foods
Manufacture of metal products
Processing of food from agricultural products
Recycling and disposal of waste
Manufacture of plastics
Manufacture of textile
Processing of timber, manufacture of wood, bamboo, rattan, palm, and
straw products
Manufacture of furniture
Manufacture of textile wearing apparel, footware, and caps
Manufacture of articles for culture, education, and sport activities
Manufacture of leather, fur, feather, and related products
Mining of other ores
Percent of total
99.3
93.0
92.1
69.4
68.6
53.6
44.4
44.0
36.9
28.8
28.7
26.2
20.5
18.7
16.7
16.5
12.5
12.3
12.1
11.8
11.5
10.6
10.3
8.9
8.9
8.3
8.2
6.9
5.8
5.8
5.4
3.8
2.7
2.4
2.3
1.7
1.4
1.2
0.3
0
Source: National Bureau of Statistics of China (2012b, 502–13).
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
77
private. These include ENN, China’s largest private gas company, which is
listed in Hong Kong and majority owned by Wang Yusuo and his wife. China
Gas, the second largest private gas company, is a Hong Kong listed company
majority owned by Liu Minghui, another private entrepreneur.17 China Gas has
concessions to supply piped gas in 160 Chinese cities and also owns a number
of long-distance natural gas pipelines and a network of compressed natural
gas refi lling stations for vehicles. The growing role of private gas companies is
strongly reflected in their share of investment in the industry, which by 2012
had risen to 40 percent, with an additional 6 percent coming from foreign
private-controlled companies (National Bureau of Statistics of China 2013c,
170–71).
The three national oil companies have retreated from another downstream activity, refining. Their once virtual monopoly has gradually eroded so
that by 2011 they accounted for less than 70 percent of refined petroleum, with
the balance roughly split between private and foreign firms (National Bureau
of Statistics of China 2012b, 502–503, 512–13, 532–33).
The state has also ceded considerable space to private firms in water
supply, a utility that is state owned in many market economies. In 1999 state
companies accounted for 90 percent of the value of output of the sector
(National Bureau of Statistics of China 2000, 414, 424). As shown in table 3.6,
by 2011 this share had fallen to under 70 percent. Foreign companies, which
were allowed to enter the sector in 2002, currently account for about a fifth
of output and registered private domestic companies another 6 percent. The
single biggest foreign player is the French firm Veoli. It has full service water
concessions in the Pudong business district in Shanghai, in several districts
in the city of Tianjin, and in other cities, where it provides water to 27 million
inhabitants.
The long-term retreat of state firms in Chinese industry did not slow
during the global financial crisis and its aftermath, a period many studies
have characterized as marking their resurgence. The average annual increase
in industrial value added in the four years after 2008 across firms of all ownership types was 12.7 percent. But the universe of state and state-controlled
firms continued to underperform, expanding by an annual average of only 9.2
percent, while registered private firms grew at an average rate of 18.2 percent
(National Bureau of Statistics of China 2010c; 2011b; 2012c; 2013a).18
17. While China Gas is usually considered a privately controlled company, Sinopec owns a 6
percent stake in the firm, and an arm of the Beijing municipal government, Beijing Enterprise
Holdings Limited, holds a stake just under that of the firm’s founder Liu Minghui. Foreign shareholders in China Gas include Korea’s SK Group and Fortune Oil PLC, a UK company.
18. The release of industrial value-added data is asymmetric in the sense that the reported growth
rate for state firms is based on a measure of output that is inclusive of both traditional state-owned
companies and companies of all registration types where the state is the majority or dominant
owner, while the reported growth rate for private firms is based strictly on the output of private
registered firms.
78
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
There has been a similar long-term decline in the importance of state firms
in the construction industry, which accounted for 7 percent of GDP in 2011
(National Bureau of Statistics of China 2013c, 54). In 1980 state firms employed three-quarters of all workers in the industry and were responsible for
just over three-quarters of the value of construction. Collective construction
enterprises accounted for the remainder of both employment and the value
of construction (National Bureau of Statistics of China 2012b, 567). There
was no private construction industry in the early years of the reform era. By
2010 the share of output of state construction firms had shrunk to less than a
third.19 The role of collective enterprises in the construction industry basically
collapsed; by 2010 they accounted for less than 4 percent of the output value
of the construction industry. In the same year foreign firms accounted for only
1 percent of output value (National Bureau of Statistics of China 2012b, 567).
Thus privately controlled domestic firms now dominate the industry: in 2010
they accounted for about three-quarters of the employment and two-thirds of
the value of output.
Services
Measuring the changing contributions of state and private firms to output
and employment in the service sector is a more difficult task. For one, the
data are much less detailed than those for industry. Although the National
Bureau of Statistics publishes annual data on aggregate service output, there
is no systematic breakdown of ownership types of service firms. Thus, disaggregation is problematic. One is forced to look at each of the 14 components
of China’s service sector. For a few, good data are available. For others, there
is useful qualitative information. But for some components the empirical
basis for analysis is much more limited. Of course, any discussion of services
has to begin by recognizing that state-owned and state-controlled firms completely dominate several key components of the sector: finance, which includes
banking, insurance, securities, and asset management; basic telecommunications; and some forms of transportation (airlines, rail, and ocean shipping),
which frequently but not always are privately run in a market economy. One
should also recognize that state units still dominate other services—educa19. Data on the number of state-owned firms, the number of workers, and the value of output of
these firms in the construction industry published by the National Bureau of Statistics cover only
traditional state-owned companies. These state-owned construction enterprises in 2010 employed
5,769,000 workers and accounted for 18.9 percent of the output in the sector (National Bureau
of Statistics of China 2012b, 567). The combined employment of state-owned construction enterprises and state-controlled construction companies in 2010 was 8,959,000 workers, according to
“An Analysis of Statistics on the Development of the Construction Industry,” Zhongguo jianshe bao
(China Construction Newspaper), May 28, 2011. Available at www.chinajsb.cn (accessed on January
30, 2013). Assuming that labor productivity in the two types of state firms is the same, stateowned enterprises and state-controlled companies together account for 29 percent of construction output.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
79
tion, health, social security, and social welfare—that are mostly government
run even in market economies. In short, unlike in agriculture, industry, and
construction, in large swaths of China’s service sector there has been no ownership transformation.
Wholesale and retail trade is the single largest component of the service
sector, accounting in 2011 for just over one-fifth of value added in services or 5
percent of GDP (National Bureau of Statistics of China 2012b, 44, 49). In 1978
wholesale and retail trade was overwhelmingly controlled by state and collective enterprises. In retailing, for example, the private sector (entirely individual
businesses at that time) was almost invisible. In 1978 only about 10 percent
of all retail units were private, and they accounted for only a tiny 3 percent of
retail employment and a minuscule 0.1 percent of retail sales. The shares of
state retail establishments, employment, and sales were 34 percent, 58 percent,
and 90 percent, respectively (National Bureau of Statistics of China 1986, 414,
445–46). The balance of retailing activity was controlled by collective enterprises, which typically meant some degree of local government ownership and
control.
By 2008 four-fifths of all retailers were privately controlled, and they
accounted for half of retail sales and a little more than half of profits in the
retail sector.20 In addition, foreign retailers, a group that includes the likes of
Wal-Mart, Tesco, Carrefour, and Metro, accounted for almost 10 percent of
retail sales. Since these firms are wholly owned by foreign private firms, they
certainly should be regarded as private. State retailers in 2008 accounted for
3.5 percent, 22 percent, and 22 percent, respectively, of retail units, employment, and sales (National Bureau of Statistics of China 2010a). In short, in
three decades the share of retail sales by state firms fell from nine-tenths to
two-tenths.
The transformation of ownership in wholesaling has not been as dramatic,
presumably because wholesale businesses on average are much larger and
raising the capital required to finance the larger inventories of these firms
was a greater challenge for private firms. In 1978 state and collective units
accounted for 98 percent of wholesale transactions. The share of wholesale
turnover accounted for by individual businesses was only 0.1 percent (National
Bureau of Statistics of China 1985, 465). By 2008 state-owned and statecontrolled and collectively owned and collectively controlled wholesaling units
together accounted for only two-fifths of sales, while privately controlled firms
20. The end point of the analysis is 2008, the year of the last economic census. Annual data on
retailing published by the National Bureau of Statistics have two shortcomings. First, the annual
data cover only above-scale retail units, which leave out many private small-scale retailers. Second,
the data are based entirely on registration status so they do not identify, for example, employment
in and sales by limited liability companies in which the majority or dominant owner is private. For
example, four-fifths of private retail units in 2008 were privately controlled. Of these 80 percent, 75
percentage points were registered private companies and 5 percentage points were firms of other
types of ownership registration in which the dominant owner was private.
80
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
accounted for 45 percent. Adding wholesaling activity controlled by private
foreign firms boosts the private total to just over 50 percent of sales (National
Bureau of Statistics of China 2010a).
The ownership transformation in the catering industry is similar to that
in retailing, presumably because there are no significant regulatory barriers
to entry and the capital required to open a restaurant is similarly modest.
On the eve of reform state and collective enterprises dominated. Almost a
third of catering firms were private, but they must have been very small on
average since they accounted for only 7 percent of employment in the industry
(National Bureau of Statistics of China 1986, 414).21 China’s 2008 economic
census shows an industry in which the structure of ownership has been totally transformed, with four-fifths of employment and almost two-thirds of
revenue generated by private catering firms (National Bureau of Statistics of
China 2010a).
It is likely that the share of employment and sales by state firms in retailing,
wholesaling, and catering has continued to decline since 2008, but measurement of this decline will have to await the results of China’s 2013 economic
census.22
In summary, in the contestable sectors of the economy where there are relatively low barriers to entry, privately registered and privately controlled companies have largely displaced state companies. This displacement includes most
of manufacturing and mining as well as construction, retailing, wholesaling,
and catering (Anderson 2012). In sectors where there are barriers to entry due
to natural monopoly, high capital requirements, or government regulation,
the dominant position state firms enjoyed in 1978 has in most cases eroded
only slightly. The best examples of the latter in industry are tobacco manufacturing, electric power, and oil and gas extraction, where state firms in 2011 still
accounted for 99, 93, and 92 percent of output, respectively (table 3.6).
The state exercises complete control of the tobacco industry through the
aptly named State Tobacco Monopoly Administration, which directly controls
the China National Tobacco Corporation (CNTC), which in turn controls cigarette companies accounting for 96 percent of the assets in the industry. The
state is unwilling to cede control of the tobacco industry for fiscal reasons—
CNTC is China’s fourth largest company in terms of net profit (Li 2012, 3–7).
The state monopolizes the oil and gas extraction industry, presumably
for strategic reasons, a pattern observed in a number of market economies as
well. Obvious examples are the state-controlled firms Pemex in Mexico and
Petrobras in Brazil. Electric power generation and distribution is a natural
monopoly. In many market economies this is a privately run, state-regulated
industry, but China has chosen direct state ownership instead.
21. Unlike retailing, catering revenue disaggregated by ownership is not available for the late 1970s.
22. Based on the lag last time, publication of preliminary results of the 2013 census is likely to
occur at the end of 2014.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
81
In services the state firms still exercise virtual monopoly control of financial services, large swaths of transportation services, and basic telecommunication services, which in many but not all market economies are provided by
private companies. Of course, the state dominates education, health, and other
social services that in many other countries also are provided by governments.
The state also plays an outsized role in journalism and publishing, broadcasting and movies, culture and art, and sports, a pattern typical in one-party,
authoritarian political systems.23
Role of Private Firms in Generating Employment and
Exports
Despite the predominance of state firms in production in certain sectors, state
firms as a group have not been a significant source of growth of employment
in the reform era. Indeed, employment in state firms has fallen in absolute
terms for more than a decade and now accounts for only 13 percent of urban
employment. Conversely, the importance of private and individual businesses
in generating employment is difficult to overstate. The analysis that follows
looks first at urban and then rural employment.
In 1978 employment in individual businesses in urban areas had been
reduced to only 150,000, down from 1.71 million in 1965, just before the
Cultural Revolution began (National Bureau of Statistics of China 1986, 92).
As shown in table 3.7, employment in individual businesses in urban China
subsequently grew rapidly, reaching 4.5 million by 1985 and 56 million by
2012. Registered private businesses in urban areas were not important until
after the legal framework for them was established in 1988. Employment
in these businesses then expanded rapidly, reaching 76 million by 2012. By
2012 individual businesses and registered private firms thus employed over
132 million individuals, almost four and a half times the number employed in
2000 and accounting for over one-third of urban employment.
But due to limitations in the methodology used by the Ministry of Human
Resources and Social Security and the National Bureau of Statistics to compile
employment data, 132 million is an understatement of employment in urban
private businesses. The first limitation is that official annual data on urban
private employment are compiled strictly on the basis of the registration status
of firms. The data do not include those employed in firms registered as limited
liability companies, shareholding limited companies, collectives, and other
registration categories in which the majority or dominant owner is private.
While the statistical authorities, as noted above, have started to provide certain
types of data based on the nature of the majority or dominant owner, annual
23. One measure of state domination in these fields is the share of investment undertaken by state
and state-controlled units. This investment ranges from a high of 71 percent in journalism and
publishing to a low of 59 percent in culture and art (National Bureau of Statistics of China 2012b,
176–79).
82
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
Table 3.7
Year
1978
1979
1980
1985
1990
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Urban registered private enterprise and individual business
employment, 1978–2012
Total
employment
(thousands)
95,140
99,990
105,250
128,080
170,410
231,510
241,230
251,590
262,300
272,930
283,890
296,300
309,530
321,030
333,220
346,870
359,140
371,020
Private
enterprises
(thousands)
n.a.
n.a.
n.a.
n.a.
570
12,679
15,268
19,987
25,452
29,937
34,584
39,543
45,810
51,240
55,443
60,709
69,120
75,570
Individual
businesses
(thousands)
150
320
810
4,500
6,140
21,361
21,312
22,688
23,770
25,212
27,777
30,125
33,100
36,090
42,445
44,675
52,270
56,430
Private enterprises
and individual
businesses
as percent of
total employment
0.2
0.3
0.8
3.5
3.9
14.7
15.2
17.0
18.8
20.2
22.0
23.5
25.5
27.2
29.4
30.4
33.8
35.6
n.a. = not applicable
Sources: National Bureau of Statistics of China (2013c, 121); ISI Emerging Markets, CEIC Database.
data on this basis have not yet been published except in the case of state firms.
In 2011 almost 75 million urban workers were employed in registration categories other than state, private, and individual (National Bureau of Statistics
of China 2012b, 125). We estimate private employment in firms of other types
based on the 2007 Industrial Survey Firm-Level Database, which enumerated employment in industrial firms of every registration category where the
majority or dominant owner is private. Assuming the share of employment
in each of these categories in 2011 is the same as that for 2007, I estimate
private employment in registration categories other than private or individual
businesses accounts for an additional 51 million workers.24 This brings the
estimated total urban private employment in 2011 to 183 million, raising the
24. This assumption probably leads to a downward-biased estimate of private employment in
firm registration categories other than private or individual businesses because the share of firms
in each of these categories where the majority or dominant owner is private has been gradually
increasing.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
83
private share of urban employment by 14 percentage points to equal half of
urban employment.
The second limitation of official employment data is that they do not
provide any information on the registration status of firms accounting for a
large share of urban employment. For example, in 2011, reported total urban
employment was 359 million, yet the statistical authorities have provided no
information on the registration status of firms employing 97 million of these
workers (National Bureau of Statistics of China 2012b, 125). This large gap
arises because the National Bureau of Statistics estimates total urban employment based on the most recent national population census and an annual labor
force sample survey, whereas data on employment by registration status are
based on reporting by enterprise units.25 But the employment data reported by
these enterprise units do not include informal urban workers—mostly migrant
workers but also some urban residents who work in these units but do not
have stable labor contracts, do not make contributions to various social insurance funds, and do not receive social benefits (Cai and Chan 2009). The latter
group includes student interns, part-time student employees, and temporary
and seasonal workers. It seems likely that the majority of these 97 million are
employed in firms in which the majority or dominant owner is private.26
This analysis supports the conclusion that almost all of the growth of
urban employment in China since 1978 is due to the expansion of private firms,
including privately owned foreign firms. Private urban employment expanded
from 150,000 in 1978, or 0.2 percent of urban employment, to an estimated
253 million in 2011 (table 3.8). In 2011 employment in privately controlled
firms accounted for two-thirds of China’s urban labor force. More impressively, the increase in private urban employment, almost entirely the result
of the formation of new privately owned or privately controlled businesses,
25. Data on self-employment and employment by private enterprises published by the National
Bureau of Statistics of China are provided by the State Administration for Industry and Commerce.
26. This assumption is based on data on the distribution of employment of migrant workers across
various industries as reported in “Report on the 2012 National Migrant Worker Investigation
Survey,” May 27, 2013, available at www.stats.gov.cn (accessed on May 28, 2013). In 2011, onequarter were employed in wholesaling and retailing, hotels and catering, and household services—all
sectors where private firms or households are dominant. About one-third were employed in manufacturing, presumably almost entirely in wholly foreign-owned enterprises (and thus private) that
rely heavily on female migrant workers to produce export goods. Just under a fifth were employed in
construction, where the role of the state is somewhat larger. The smallest share, less than 7 percent,
was employed in transport, storage, and post, a sector that the state still heavily dominates. But this
domination is due to the state’s sole ownership of the national rail system and majority ownership
of the largest airlines and international shipping companies. It is likely that migrant workers were
employed in trucking and logistics businesses, where private ownership is more important. Based
on this analysis, I assume that just over 80 percent of the 97 million urban workers employed in
firms where the registration status is not provided were employed in private firms.
84
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
Table 3.8
Urban employment in privately controlled firms, 2011
Enterprise type
Registered urban private enterprises
Urban individual businesses
Registered enterprises other than private where the majority or
dominant owner is private
Unclassified enterprises that are majority private
Total
Thousands of employees
68,959 a
52,270
51,312
80,000
252,541
a. This number is very slightly less than the 69.12 million shown in table 3.7 because of a tiny number of registered private enterprises in which the majority or dominant owner is not private.
Sources: Author’s calculations based on data from the National Bureau of Statistics, Industrial Survey Firm-Level
Database (2007); National Bureau of Statistics of China (2012b, 125).
accounts for 95 percent of the growth of the urban labor force since 1978.27
Government employment also expanded by an estimated 15 million over the
same period.28 The corollary of these changes is that the absolute numbers of
employees in state and collective firms in urban areas today is slightly less than
in 1978.29 But, given the large expansion of the urban labor force, the employment share of state and collective firms fell from 99.8 percent in 1978 to about
18 percent by 2011.
The growth of employment in rural China is also almost entirely attributable to the private sector. As noted in chapter 1, agricultural activities were
essentially privatized in the early 1980s. Given large increases in labor productivity in agriculture, farm employment is now slightly less than it was in 1978.30
But by 2011 there was a more than offsetting growth of 60 million workers in
registered private enterprises and individual business in rural areas (National
Bureau of Statistics of China 2012b, 125).
There is one potential counterindicator to the finding that expanding
employment of private firms accounts for virtually all of the growth of China’s
labor force since 1978: As late as 2010 China’s 27.4 million township and
village enterprises still employed 160 million workers that were classified as
27. The total urban labor force increased by 264 million from 95.14 million in 1978 to 359.14
million in 2011 (National Bureau of Statistics of China 2012b, 126).
28. This is a crude estimate based on the data in table 4.2, which shows that employment in the
government and party bureaucracy expanded by roughly 5 million, or 11 percent, between 1999
and 2011. I assume the same pace of increase in the 1980s and 1990s.
29. The sum of the increase in employment in private firms, 252.39 million, and the increase in
government employment, 15 million, is 267.4 million, which exceeds the 264 million increase in
the total labor force. Thus employment in state enterprises, including traditional state-owned
firms and firms of other registration types where the state is the majority or dominant owner,
must have declined slightly.
30. Employment in what the statistical authorities classify as primary industry fell from 284
million in 1978 to 266 million in 2012 (National Bureau of Statistics of China 1990, 114; 2012,
125). Primary industry includes forestry, animal husbandry, and fisheries as well as agriculture.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
85
rural (National Bureau of Statistics of China 2011a, 109). The collective origins
of most of these firms might give the impression that local governments are
still playing a role in generating nonagricultural jobs in the countryside. But
from the mid-1980s onward, a growing share of employment in township and
village enterprises was in firms that were either individual businesses or registered private firms. After the mid-1980s employment in collective township
and village enterprises grew slowly, peaking in 1995 and then falling in the
ensuing decade (Naughton 2007, 286–87). By 2010, 95 percent of these 27.4
million rural production units were registered as either private enterprises
or individual businesses. These private and individual businesses employed
almost 120 million workers, about three-quarters of the total number of
workers in township and village enterprises (Zhang Tianzuo 2011, 137).
In addition, there were about 1.2 million township and village enterprises
registered as share cooperative enterprises, limited liability companies, shareholding limited companies, or foreign enterprises; these enterprises employed
38 million workers. It is likely that the share of these firms that are majority
privately owned is quite a bit higher than the share of these firms in urban
industry and that most of their employees should be included in any estimate
of private employment.31 Collective township and village enterprises, the
universe that is most likely to continue to have significant local government
ownership and control, accounted for only 0.4 percent of all township and
village enterprises and employed only 4 million workers in 2011. Including
both those employed in township and village enterprises that are registered
as individual or private businesses and an estimated 25 million employed by
firms in other registration categories that are privately owned raises private
sector employment in private township and village enterprises to 145 million
in 2010. Thus farm employment appears to have shrunk, and private nonagricultural firms account for virtually all the growth of employment in rural
China since 1978.
The growing role of the private sector in generating China’s exports is
also worth noting. As reflected in figure 3.3, state-owned companies accounted
for two-thirds of China’s total exports as recently as 1995. But their share fell
continuously starting in the second half of the 1990s. By 2012 state firms
accounted for only 11 percent of China’s by then much larger volume of
exports. Initially, as the role of state-owned firms in exporting receded, foreign
firms expanded their share of China’s exports, allowing China’s total exports to
continue to grow robustly. But the foreign firm share of total exports peaked in
31. For example, in 2007 almost 70 percent of the employees in above-scale industrial enterprises
registered as limited liability companies worked in firms where the majority or dominant owner
was private. This share is likely to be considerably higher in township and village enterprises registered as limited liability companies because that universe includes many firms that in the industrial sector would be classified as below-scale and thus much more likely to be privately owned.
Moreover, average employment in township and village enterprises was less than six workers, a
scale so small that almost all the registration categories comprise firms in which the sole or dominant owner is private (Zhang Tianzuo 2011, 137).
86
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
Figure 3.3
Sources of Chinese exports by ownership status, 1995–2013
percent share of total exports
80
70
60
50
47
40
39
30
RISE OF THE PRIVATE SECTOR
20
11
10
0
1995
1996
1997
1998
1999
2000
2001
2002
State
2003
2004
2005
Foreign
2006
2007
2008
2009
2010
2011
2012
2013
Private
Notes: Prior to 2013, exports by privately owned firms are estimated as the difference between total exports and the sum of exports by state, foreign, and collective firms. Collective
enterprise exports are not shown. In 2013, collective enterprise exports were only 3 percent of total exports.
87
Sources: General Customs Administration, www.customs.gov.cn; ISI Emerging Markets, CEIC Database.
© Peterson Institute for International Economics | www.piie.com
2005, and China has since become increasingly dependent on private domestic
firms to sustain its role as a major exporting economy. Private firms’ share
of exports has grown so much that since 2009 the expansion of the value of
private firms’ exports has exceeded that of foreign firms.32
Since 95 percent of China’s exports are manufactured goods, the rise of
private firms as important exporters reflects their displacement of state firms
in most branches of manufacturing, as detailed earlier. The rise of private
firms in exporting also explains the evolution of China’s export structure.
The share of processed exports, which involve the assembly of imported parts
and components largely carried out by foreign firms, is declining, resulting
in an increase in the value-added component of China’s exports. The rise of
the private sector has led China to move up the value chain in both industrial
production and exports.33
While it is little noticed, private firms are also becoming a factor in China’s
growing outbound foreign direct investment. Large-scale investments abroad
by giant state-owned companies, such as the China National Offshore Oil
Corporation (CNOOC), capture inordinate attention in the western press.
CNOOC’s failed $18.5 billion bid to take over US oil company Unocal in
2005 and the firm’s successful acquisition of Canadian energy firm Nexen in
2013 for $15.1 billion are just two examples. But China’s private firms are also
investing abroad. The earliest notable example of an outbound investment
into the United States was Lenovo’s acquisition of the personal computer business of IBM in 2005. More recent examples include Wanda Group, controlled
by one of China’s richest entrepreneurs. It paid $2.6 billion in 2012 to acquire
AMC Entertainment, which operates the second largest movie theater chain
in the United States. Privately controlled Wanxiang, China’s largest auto parts
company, entered the US market in 1994 and now has manufacturing operations in 14 states employing 6,000 workers. In 2013 in a widely publicized
transaction Wanxiang America, the Chicago-based arm of Wanxiang, acquired
the US firm A123 Systems, a Massachusetts-based maker of electric car
batteries, for $257 million.34 In China’s largest acquisition in the United States
to date, Shuanghui International, a private company, acquired Smithfield
Foods in 2013 for $7.1 billion. Overall, private Chinese companies accounted
for almost two-fifths of cumulative Chinese direct investment in the United
States from 2000 through the second quarter of 2012.35 Private Chinese firms
32. For example, in 2013, exports of private firms increased by $144.3 billion, while exports of
foreign firms rose only $21.5 billion. Exports of state firms in 2013 fell by $7.3 billion.
33. Louis Kuijs, “Is China upgrading its industrial structure?” May 21, 2013, RBS Emerging
Markets Asia Top View/China.
34. Zhang Yuwei, “US Approves Wanxiang’s $257 million purchase,” China Daily, January 30, 2013,
p. 1.
35. Rhodium Group, China Investment Monitor. Available at http://rhgroup.net/interactive/
china-investment-monitor (accessed on October 5, 2012).
88
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
are also actively investing in Europe, where they accounted for almost 30
percent of a cumulative $15 billion in Chinese investment from 2000 through
2011 (Hanemann and Rosen 2012, 46). However, in the aggregate, state firms
still dominate China’s global outward foreign direct investment, in large part
because investments in natural resources and raw materials (almost all outside
the United States and Europe) loom very large in these outflows and state
firms most frequently generate them.
Explaining the Growth of the Private Sector
So far, this chapter has made the case for the continued overall rise in the
role of the private sector in China and established why the nature of the data
have led some analysts to underestimate its role. But what explains the private
sector’s steady displacement of state firms in manufacturing and its significant
inroads in some other parts of the economy? The balance of the chapter argues
that the evolution of state policy toward the private sector and the greater
efficiency of private firms explain the shift. In the early years of reform the
policy environment for private nonagricultural business was basically hostile,
but very gradually this changed, facilitating the emergence of an increasingly
robust private sector. But the higher efficiency of private businesses compared
with state firms was also critical to their displacement of state firms as major
sources of output across broad swaths of the economy as well as their ability
to generate almost all the growth of employment since 1978. Greater efficiency
of private firms is reflected in a much higher return on assets, meaning that
they have a higher level of retained earnings relative to their assets than state
companies do. Higher retained earnings allow private firms to expand their
assets more rapidly, which, when combined with higher returns, leads to much
more rapid growth than state firms enjoy.
Evolution of State Policy toward the Private Sector
As already noted, government regulations issued early in the reform era did
allow the emergence of individual businesses in urban and rural areas. For
example, the National People’s Congress in March 1978 adopted a constitutional amendment allowing “individual laborers” to operate “within the limits
permitted by law” (Tsai 2007, 50). The limits included a provision that these
family businesses could not employ more than seven nonfamily members. The
State Council followed up in 1981 with detailed regulations governing individual businesses in urban areas and issued similar regulations for rural areas
three years later.
However, in the 1980s and the early part of the 1990s the environment for
private firms was generally inhospitable. This was not surprising in a country
where the ruling party for decades had emphasized that private entrepreneurs
were exploiters of the working class. The government did not promulgate the
Provisional Regulations on Private Enterprises until 1988, a full decade into
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
89
the economic reform era.36 These regulations lifted an important constraint
on the development of the private sector because they provided a legal framework for private firms with eight or more employees. However, the regulations
provided only for private sole proprietorship enterprises, which meant that
an entrepreneur’s personal wealth could not be legally separated from the
assets of his or her business. The absence of limited liability in sole proprietorships naturally tended to restrict the size of these firms; provisions for limited
liability companies and shareholding limited companies did not come into
effect until 1994. Moreover, in the 1980s private firms were generally regarded
as illegitimate and were vulnerable to special local government levies, which
many private firms tried to avoid by registering as shareholding cooperatives
or collectives. But this required them to pay a management fee to the local
government and subjected them to the requirement to contribute to the local
government’s accumulation fund (Nee and Opper 2012, 110–11).
Even into the mid-1990s there was substantial policy discrimination
against private firms. For example, as explained in chapter 2, the authorities
saw the Company Law of 1994 as a key method for transforming traditional
state-owned enterprises into corporations. The law established high minimum
capital requirements, designed in part to prevent private enterprises from
becoming companies and thus enjoying limited liability. The state set the
minimum capital requirement for most production enterprises to register as a
limited liability company at RMB500,000 and at RMB10,000,000 to register as
a shareholding limited company. For comparison, the average worker annual
wage was then RMB4,500, and bank savings, the most readily available source
of funds for an entrepreneur, averaged only RMB1,800 per capita (National
Bureau of Statistics of China 1995, 113, 257).
Despite the apparently very high hurdle posed by these minimum capital
requirements, private firms quickly adopted the same legal structures as transformed state companies—that is, limited liability companies and joint stock
companies—and they frequently did so by short-term borrowing to meet the
legal minimum capital requirement and repaying after registration. The incentive for entrepreneurs to take advantage of these new legal forms of ownership
was more than limited liability. “[P]rivate businesses registered as sole proprietorships are easily identifiable by the company’s name, but firms registered as
an LLC or JSC could be anything from wholly state-owned to wholly privatelyowned companies” (Nee and Opper 2012, 114). Bureaucrats could not easily
identify private limited liability and joint stock companies and thus private
firms were protected from discrimination. (Unfortunately, this protection
has also made tracking the growth of the private sector more difficult for the
analyst.)
After the Company Law took effect, there was a massive change in the
recorded ownership structure of Chinese firms. First, the number of collec36. State Council, “Provisional Regulations on Private Enterprises,” June 25, 1988. Available at
www.ynfzb.cn/Finance/ChuangYeGS/201107140960.shtml (accessed on October 15, 2013).
90
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
tive enterprises declined rapidly, as these firms reregistered to reflect their true
private ownership, a process sometimes referred to as “taking off the red hat.”
The number of collective enterprises (not just in industry but across all sectors)
fell from 1.5 million in 1996, when they accounted for just under three-fifths of
all firms, to only 265,000 in 2004, when they accounted for only 4 percent of all
firms. Over the same period the number of private firms soared from 443,000
to 3.6 million, reflecting both the creation of private firms and the reregistration of collective firms as private (National Bureau of Statistics of China 1998;
State Council Leading Small Group Office on the Second National Economic
Census and the National Bureau of Statistics of China 2009).37
Second, the ownership arrangements within the universe of registered
private firms were also transformed. Before 1994 all registered private firms
were sole proprietorship enterprises or private partnership enterprises, but
only three years later, 48 percent of private firms were registered as limited
liability companies, a share that rose to 65 percent by 2004 (Nee and Opper
2012, 116).
The government took other steps to promote the legitimacy of the private
sector. In March 1999 the Ninth People’s Congress approved a constitutional
amendment identifying the nonstate economy as “an essential component” of
a mixed economy, a clear improvement from its previous designation as “an
important component” of a state-dominated economy (OECD 2000, 52). Two
years later Party General Secretary Jiang Zemin, in a speech marking the 80th
anniversary of the founding of the Chinese Communist Party, in effect invited
private entrepreneurs to join the party. In a followup a year later, the party
clarified that it would no longer discriminate against private entrepreneurs
but would embrace them because of their contributions to China’s economic
development (Nee and Opper 2012, 64).
The state also sought to expand the scope of businesses open to private
firms by gradually reducing restrictions on entry in various sectors and taking
other steps that made the expansion of private activities possible. The State
Council launched this initiative in 2005 with a directive that came to be known
as the 36 Articles (State Council 2005). It specifically called for encouraging
private investment in a number of sectors previously reserved exclusively for
state firms including electric power, telecommunications, railroads, civil aviation, and petroleum. And it encouraged all financial institutions to raise the
share of their loans to nonstate enterprises. City commercial banks, city credit
cooperatives, and policy banks were specifically advised to provide more credit
and a broader range of financial products and services to farmers, household
businesses, medium and small-scale enterprises, and nonstate enterprises. The
banking regulator, the China Banking Regulatory Commission, followed up a
37. These data are based on the number of enterprise legal persons.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
91
few months later with specific guidelines on increasing loans to and providing
enhanced financial services to small enterprises.38
Revisions of the Company Law in 2006 improved the situation for private
firms in two respects and led to a further increase in the share of private firms
that were organized as limited liability companies. First, the minimum capital
requirements to register as a limited liability company or shareholding limited
company were reduced to RMB30,000 and RMB5,000,000, respectively, onetenth and one-half their previous levels. Second, the revision allowed singleperson limited liability firms for the first time. Previously, a single owner of a
registered private company that sought the advantage of the limited liability
legal structure had to circumvent the provisions of the 1994 law by the use of
dummy shareholders so he or she could qualify as a private limited liability
company.
Progress in opening up various sectors to private investment, however,
was slow. For example, following the promulgation of the 36 Articles in 2005,
entrepreneurs were able to establish private airline companies such as Spring,
Juneyao, Okay, and East Star. But the new entrants were handicapped by state
control on ticket prices, which limited their ability to gain market share, and
the government initially would not grant them plum routes, such as ShanghaiBeijing, which were monopolized by incumbent state airlines such as Air China,
Shanghai Airlines, and China Eastern.39 Moreover, approval of new private
airlines was suspended after July 2007 and did not resume until May 2013.40
Partly as a result of the slow progress of private firms in overcoming regulatory obstacles, even in sectors that formally had been opened for private
investment, the State Council issued follow-on guidelines in May 2010 further
liberalizing investment by nonstate firms.41 Since this document also had
36 articles, it is widely known as the “New 36 Articles.” In this document the
State Council called for thorough implementation of the 36 articles of 2005
and newly opened a number of sectors for private investment. These included
certain types of infrastructure, public housing, certain types of financial
services, and the fields of culture, tourism, and sports.
In a further liberalizing step, in 2011 the government lifted the restriction on the maximum number of nonfamily members that can be employed in
individual businesses, so the growth of these firms over time may be less con-
38. China Banking Regulatory Commission, “Guiding Opinions on Expanding the Small
Enterprise Loan Business of Banks,” July 15, 2005. Available at www.cbrc.gov.cn (accessed on
October 17, 2006).
39. Wang Xu, “Slowly, Private Firms Chisel an Investment Wall,” Caing, April 20, 2010. Available at
http://english.caing.com (accessed on June 25, 2010).
40. Simon Rabinovitch, “Frills airline reflects challenge to Chinese carriers,” Financial Times, June
17, 2013, p. 7.
41. State Council, “Several Opinions on Encouraging and Guiding the Healthy Development of
Non-state Investment,” May 7, 2010. Available at www.gov.cn (accessed on June 25, 2010).
92
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
strained and they may provide even more employment opportunities (Lardy
2012, 34).
The government in recent years also has used tax policy to promote the
growth of more than 6 million small and microenterprises, which are overwhelmingly registered private companies, and individual businesses. In 2008
the government introduced a corporate income tax rate of only 20 percent,
rather than the standard 25 percent for small, microprofit enterprises.42 In
2010 the government extended this initiative by introducing an even lower rate
of 10 percent for firms with annual taxable income of RMB30,000 or less. In
2011 this low rate was applied to firms with income up to RMB60,000, and
in 2014 the level was raised further to RMB100,000.43 In addition, beginning
in August 2013 the government exempted both small and microenterprises
as well as individual businesses with annual sales revenue below RMB240,000
from both value-added taxes and business taxes.44 Finally, beginning in 2014,
2.3 million service sector enterprises began to pay a 3 percent value-added tax
instead of a 5 percent business tax on their top-line revenue. These initiatives
have reduced taxes for some 10 million small and microenterprises and individual businesses (State Council 2014).45
42. The criteria for microprofit enterprises have been set by the tax authorities and used only to
determine tax obligations of firms. In industry, for example, a microprofit enterprise must have
before-tax earnings of no more than RMB300,000, no more than 100 employees, and assets of no
more than RMB30 million. These criteria are set forth at www.chinatax.gov.cn (accessed on May
28, 2014).
43. “Tax cuts for small firms,” China Daily, April 10, 2014, p. 13. Individual businesses are not
subject to the corporate income tax but do pay value-added taxes and business taxes. In addition,
owners of individual businesses are subject to personal income tax on the income they derive from
their businesses. The rates that these entrepreneurs pay differ from the rates imposed on wage
income.
44. The definition of microenterprises (微型企业) was introduced when the Ministry of Industry
and Information Technology, the National Bureau of Statistics, the National Development and
Reform Commission, and the Ministry of Finance jointly issued a new regulation on the classification of enterprises by size, replacing the previous 2003 regulation. “Notice concerning the
Regulations on the Standards for Classification of Medium and Small Enterprises,” June 18, 2011.
Available at www.gov.cn (accessed on August 13, 2013). Prior to the change, industrial firms were
classified as small (小型企业) if they had sales revenue of less than RMB30 million, fewer than
300 workers, or assets less than RMB40 million. After the change, industrial firms were classified as microsized if they had less than RMB3 million in sales or fewer than 20 employees and
as small if they had sales between RMB3 million and RMB20 million and between 20 and 300
workers. This reclassification, which lowered the upper limit on small-size firms and created the
new microcategory, facilitated the government’s efforts to prioritize the development of small and
microenterprises.
45. The measurement of the number of enterprises used here is hu, meaning that branches of firms
are included in the count. The 10 million includes a small amount of double counting since it is
the sum of the number of firms benefiting from a reduction in the corporate income tax, the elimination of the value-added tax and the business tax, and the switch to a 3 percent value-added tax in
place of a 5 percent business tax. Some of the 1.2 million firms eligible for the reduced corporate
income tax rate also benefited from one of other three tax initiatives.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
93
Financing the Private Sector
While the political, legal, and regulatory environment for private economic activity in China gradually improved, many news analyses, commentators, investment banks, and academics have focused on the lack of access to bank credit,
as well as to debt and equity finance, as major constraints on the development
of private firms. The Economist argued in 2012 that “state giants soak up capital and talent that might have been used better by private companies” while
the New York Times the same year said that small and medium businesses “now
receive as little as 3 percent of bank lending even as they account for at least
half the country’s economic activity.”46 Analysts at Roubini Global Economics
charge that “banks exist to provide funding for the government and its stateowned enterprises” (Wolfe and Aarsnes 2011), and western investment bankers
with lengthy experience in China argue that “Chinese banks overwhelmingly
lend to SOEs and always have” (Walter and Howie 2011, 4). James McGregor
charges that China’s largest state-owned banks “serve as ATMs for the SOEs”
(2012, 7). David Pilling, the Asia editor of the Financial Times, wrote in November
2013 that China’s state-owned enterprises, “far from shrinking have grown, recipients of massive injections of state funds designed to keep economic growth
growing.”47 Two months later the Financial Times opined in an editorial that
“the more vibrant segments of the private sector . . . are typically starved of cash.
Meanwhile, the larger state-owned enterprises, as well as local governments,
enjoy easy access to loans. . . .”48 Deutsche Bank Securities in a November 2013
report argued that the “discrepancy between private enterprises that account
for nearly 70 percent of GDP but get only 20 percent of the credit is the greatest
constraint on growth.”49
Retained Earnings
Before turning to an analysis of actual flows of bank credit, access to debt and
equity markets, and bankers’ acceptances and trust loans, which are important components of China’s growing shadow banking system, it is important
to recognize that it is retained earnings that have financed the vast majority
of investment in China in the reform era and not borrowing from banks, issuing debt, selling equity, or acquiring funds from the shadow banking system.
According to a survey by the World Bank, the ability of Chinese firms to finance
46. “The rise of state capitalism,” Economist, January 21, 2012, p. 11. Keith Bradsher, “China Pushes
Deposit Insurance in Bank Overhaul,” New York Times, December 13, 2012, p. B9.
47. David Pilling, “The ghost at China’s third plenum: demographics,” Financial Times, November
11, 2013, p. 9.
48. “China’s dangerous credit addiction,” Financial Times, January 16, 2014, p. 8.
49. Peter Hooper, Gilles Moec, Michael Spencer, and Torsten Slok, “China’s final stage of reform,”
November 8, 2013, Deutsche Bank Research Global Economic Perspectives.
94
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
investment from retained earnings is substantially higher than in other uppermiddle-income countries.50 The heavy reliance on internal finance as a source
of investment is a significant change from the pre-reform era, when traditional
state-owned enterprises dominated the economy and were required to turn over
almost all of their profits to the Ministry of Finance. These funds were then
used to finance investment through the state budget according to priorities
established by the State Planning Commission. In 1978 more than three-fifths
of investment in state-owned units was financed through the state budget. But
one of the most important reforms undertaken after 1978 was to allow firms to
retain most of their after-tax profits. By 1997 less than 3 percent of industrial
investment was financed through the state budget (Naughton 2007, 304).
The important and growing role of retained earnings as a source of investment funds in the 1990s and 2000s is reflected in figure 3.4. From 1992
through 1999 retained earnings financed about 40 to 50 percent of the investment of all nonfinancial corporations. But from 2000 through 2008, this ratio
rose to an average of 71 percent. Even when credit growth exploded during the
2009 global financial crisis and its immediate aftermath, an average of 56 percent of all investment was financed from retained earnings. In short, retained
earnings since 2000 have consistently been a far more important source of investment finance than the sum of the funds firms borrow from both banks
and nonbank financial intermediaries, raise through the sale of stock on the
Shanghai, Hong Kong, and other global equity markets, and raise through the
sale of bonds and other debt instruments both at home and abroad. Thus the
relative financial performance of state versus private firms may go a long way
toward explaining the differential growth performance, analyzed earlier in this
chapter, of these two types of firms.
Many studies have addressed the relative productivity performance of
state and private firms in China’s industrial sector. Regardless of whether
ownership is defined by registration status or the nature of the majority or
dominant owner, these studies have universally found that total factor productivity is substantially higher for private firms.51 The OECD found that productivity in the universe of registered private firms and privately controlled firms
in 1998–2003 was more than twice that of traditional state-owned enterprises
and more than half again as high as state firms that had converted their ownership structures to become limited liability companies or shareholding limited
companies (OECD 2005, 86). Similarly, a study for the much longer 1978–
2004 period found that total factor productivity growth in the state sector was
only 1.7 percent per annum, only two-fifths the 4.3 percent per annum pace of
the nonstate sector (Brandt, Hsieh, and Zhu 2008, 694).
50. World Bank, “Enterprise Surveys: China Country Profi le 2012,” July 2013. Available at www.
enterprisesurveys.org (accessed on September 10, 2013).
51. In contrast, because state firms typically have far more capital per worker they usually have
higher output per worker than private firms. But this does not translate into higher profits per
unit of capital or to higher total factor productivity.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
95
96
Figure 3.4
MARKETS OVER MAO
percent
Retained corporate earnings as a share of corporate investment, 1992–2011
90
80
71
70
60
56
50
40
30
20
10
0
1992
1993
1994
1995
1996
1997
1998
Retained earnings/gross capital formation
1999
2000
2001
2002
2003
2004
2005
2006
Average share, 2000–2008
Note: Retained earnings are the disposable income of nonfinancial corporations reported in the flow of funds.
Sources: National Bureau of Statistics of China (2013c, 80–81); ISI Emerging Markets, CEIC Database.
© Peterson Institute for International Economics | www.piie.com
2007
2008 2009
2010
Average share, 2009–11
2011
Faster growth of factor productivity in private firms has translated into
much stronger financial performance. The OECD study found that private
firms were able to maintain their earnings at a relatively high share of value
added and that their return on assets and return on equity increased substantially between 1999 and 2003. In 2003 private companies’ average return on
both assets and equity was about half again as high as state firms’ returns
(OECD 2005, 87, 99). The World Bank also found a significantly higher return
on equity by nonstate firms during the period 1998–2009. By 2009 the gap
was huge, with nonstate firms’ returns on equity at 20 percent, running 9.9
percentage points ahead of state firms (World Bank 2012, 111). Similarly,
David Dollar and Shang-Jin Wei (2007), based on a survey of a random sample
of 12,400 firms conducted in 2002–03, found that private firms’ average return
to capital was half again as high as that of wholly state-owned firms and also
well above returns of majority and minority state-owned firms.
Data for the industrial sector going back to the mid-1990s, shown in figure
3.5, demonstrate that registered private enterprises consistently have earned a
higher return on assets than the combined universe of state-owned enterprises
and state-controlled firms.52 The reforms of the state-owned industrial sector
analyzed in chapter 2 and the increased competition as a result of China’s entry
into the World Trade Organization did lead to a very substantial improvement
in the efficiency of these firms for a decade after 1998. The return on assets in
state firms rose from a mediocre level of less than 1 percent in 1997–98 to a
peak of 6.8 percent in 2007. But even at this peak, registered private firms were
much more efficient, recording a return on assets of 9.5 percent. Since then,
with the exception of 2010, performance at state firms has fallen, hitting 4.9
percent in 2012. In contrast, except in 2009 and 2012, returns at registered
private firms have continued to rise. Thus the margin in favor of privately
registered firms has widened dramatically since the mid-2000s. By 2012 the
gap was almost three to one.
There are no time-series data that would allow the calculation of return
on assets of state and private firms in the service sector. But data in the 2008
52. Ideally, I should compare the relative productivity performance of the combined universe of state and state-controlled industrial firms with that of the combined universe of private
and privately-controlled industrial firms. But, as discussed, time-series data on the profits and
assets of privately-controlled industrial firms are not available. But there are two reasons to
believe that the productivity performance of registered private firms is a reasonable measure of
the larger universe of private firms, including those of other registration types where the majority
or dominant owner is private. First, as shown in figure 3.1, the output of registered private firms
accounted for two-fifths and three-fifths of the output of the larger universe of registered private
and privately controlled firms (in both cases including below-scale firms) in 2003 and 2007, respectively. Second, and more important, in 2007, a year for which complete data are available, the
gap between the return on assets in the narrow universe of registered private industrial firms,
9.5 percent, and the universe of industrial firms in other registration categories that are privately
controlled, 8.3 percent, is relatively small. The return for the entire universe of domestic private
firms was 8.9 percent. Thus using returns of registered private firms as a proxy for all private firms
overstates returns by 0.6 percentage points or only 7 percent (= 9.5/8.9).
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
97
98
MARKETS OVER MAO
Figure 3.5
Return on assets of state and private industrial enterprises, 1996–2012
total profits/total assets (percent)
16
14
13.2
12
10
8
6
4.9
4
2
0
1996
1997
1998
1999
2000
2001
2002
2003
2004
State
2005
2006
2007
2008
2009
2010
Private
Note: Total profits are earnings before corporate income tax.
Sources: National Bureau of Statistics of China (2013c, 475–93); National Bureau of Statistics of China, www.stats.gov.cn (accessed on February 28, 2014).
© Peterson Institute for International Economics | www.piie.com
2011
2012
economic census support two points: that the return on assets of state enterprises in the service sector is substantially below the returns that state industrial firms earn and that the performance of state firms in services is far inferior
to service firms with other forms of ownership.53
Disaggregating the state industrial and service sector provides further
insight. The average return on assets of state firms is flattered by the extraordinarily high returns earned by CNOOC, China Mobile, the China National
Tobacco Corporation, and a few other large state firms that earn supranormal
profits and thus have extraordinarily high return on assets. CNOOC, as noted
in chapter 1, enjoyed profit margins in 2004–12 averaging 44 percent and
generated an average return on assets of 24.5 percent. The profit margins of
China Mobile in 2004–12 averaged 33 percent, translating into a return on
assets that averaged 19 percent. The China National Tobacco Corporation in
2012 earned a return on assets of 14 percent (National Bureau of Statistics of
China 2012b, 512–15).54 In the universe of registered private firms, there does
not appear to be a small number of firms that make a comparable outsized
contribution to profits. Consequently, the average return on assets for private
registered firms is probably fairly close to the median return, but the median
return in the universe of state firms is likely below the average, both in services
and in industry. The significance of this will be analyzed in the next chapter.
The superior financial performance of private industrial firms—whether
that universe is defined by who actually controls the firm (as in the case of the
OECD study) or by the registration status of firms (as in the case of the World
Bank study) or the mixed comparison of state-owned and state-controlled
versus registered private (as in the case of the Chinese data on the industrial
sector reflected in figure 3.5)—must explain a large share of their relatively
rapid growth. Private firms were able to rely much more on retained earnings
to finance their expansion, partially overcoming the handicap of the more
limited access of private firms to bank credit and other external sources of
finance. However, as will be shown below, private-firm access to credit also has
improved, contributing further to their superior growth performance.
Bank Credit
The People’s Bank of China has published four data series reflecting the flow
of credit to the private sector. These series are available for different periods,
and each has specific strengths and weaknesses. No single published series
53. The data, covering 10 of the 14 subsectors of services, show that the return on assets of state
and state-controlled firms was 3.4 percent, while returns in nonstate services enterprises were
6.6 percent. The omitted subsectors are real estate; transport, storage, and post; public management, social welfare, and social organizations; and finance (National Bureau of Statistics of China
2010a).
54. The data are for 2011 for 117 state-owned tobacco companies, all of which are under the
purview of the China National Tobacco Corporation.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
99
provides a comprehensive measure of the flow of credit to the private sector
during the entire reform era.
The first three data series are partially shown in table 3.9. Data from the
first series, in the first set of columns, were cited by the Economist in its 2012
survey of state capitalism and by several academic studies in support of the
view that the private sector in China has had little access to credit in the reform
era.55 This judgment is flawed for two reasons. First, the data in the first series
cover credit flows only to what might be called the narrowly defined private
sector—that is, registered private companies and individual businesses. They
do not include lending to the large number of limited liability companies in
which the majority or dominant owner is private, which, as will be shown, is
substantial. Second, the first series includes only short-term credit.56
The second data series (middle columns of the table), which begins only
in 2002, has the same coverage as the first but includes medium and long-term
loans. In 2002–09, the years for which data in both the first and the second
series have been released, the data reveal that when medium and long-term
loans are included the share of credit going to the private sector is four to ten
times larger than the data in the first series. This sizable gap underscores why
it is a mistake to look only at short-term loans to measure the flow of credit to
the private sector.
The third set of columns shows business loans from financial institutions
to households, including working capital loans to farm households.57 This series excludes lending to registered private businesses. Farm households are not
classified as individual businesses, so their borrowing is not included in the
data in the first two series. But business loans to farm households should be
included in any attempt to measure credit flows to China’s private sector for
two reasons. First, farming in China is now almost entirely private. Second,
loans to farm households may indirectly provide credit to the individual nonagricultural businesses that employed over 80 million rural workers in 2010.58
55. The Economist writes, “In 2009 private firms accounted for only 2 percent of China’s official outstanding loans” (January 21, 2012, special report, p. 15). Kellee Tsai (2002, 2) uses this
data series to assert, “As of the end of 2000, less than 1 percent of loans from the entire national
banking system had gone to the private sector.” Ligang Song (2005, 117) cites data from the same
series in support of the view that private enterprises and individual businesses received only 0.62
percent of loans from all banks at the end of 1999.
56. In China, short-term loans are loans with a maturity of up to and including one year.
57. Loans by financial institutions to households are divided into business loans and consumer
loans. The former category is discussed in the text. Consumer loans include car loans, student loans,
credit card loans, and, most importantly, home mortgage loans. At year-end 2011, total loans by financial institutions to households stood at RMB13,607 billion, 21 percent of all loans outstanding
from the financial system. About two-thirds of these loans were consumer loans; the balance were
business loans. See table A.4 for data on consumption loans to households in recent years.
58. In 2010 there were 25 million workers employed in individual businesses in rural areas and 61
million workers in township and village enterprises in which the actual owner was an individual
(National Bureau of Statistics of China 2011a, 109; Zhang 2011, 137).
100
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
Table 3.9
Loans by financial institutions to private businesses, 1980–2013
Short-term loans to private and
individual businesses
RISE OF THE PRIVATE SECTOR
Year
1980
1985
1990
1995
2000
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Millions of
renminbi
23
1,065
4,020
19,620
65,460
105,877
146,159
208,149
218,075
266,757
350,766
422,382
712,101
n.a.
n.a.
n.a.
n.a.
Percent of
total loans
0.01
0.17
0.23
0.39
0.66
0.81
0.92
1.17
1.12
1.18
1.34
1.39
1.78
n.a.
n.a.
n.a.
n.a.
Loans to private and
individual businesses
Millions of
renminbi
n.a.
n.a.
n.a.
n.a.
n.a.
1,033,300
1,514,700
1,456,000
1,698,500
2,122,100
3,752,500
4,173,900
5,104,600
7,073,700
9,091,800
11,017,500
n.a.
Percent of
total loans
n.a.
n.a.
n.a.
n.a.
n.a.
7.39
8.92
7.72
8.21
8.91
13.51
13.04
11.99
13.89
15.62
16.37
n.a.
Loans to
household businesses
Millions of
renminbi
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
829,807
965,229
1,423,716
1,792,326
1,984,763
2,645,325
3,747,845
4,729,472
5,694,939
6,878,276
Percent of
total loans
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
4.40
4.67
5.97
6.45
6.20
6.22
7.36
8.13
8.46
9.76
n.a. = not available
Note: Financial institutions include the three policy banks, state commercial banks, shareholding commercial banks, urban commercial banks, rural commercial
banks, rural cooperative banks, urban credit cooperatives, rural credit cooperatives, trust and investment companies, foreign banks, the postal savings bank,
finance companies, and leasing companies.
101
Sources: China Banking Society (1990, 152–53; 1996, 429; 2006, 389; 2008, 402); ISI Emerging Markets, CEIC Database; People’s Bank of China, www.pbc.gov.cn
(accessed POFPebruary 28, 2014); All-China Federation of Industry and Commerce (2011, 76; 2012, 66; 2013, 68).
© Peterson Institute for International Economics | www.piie.com
When a farm household has access to working capital loans to finance the purchase of seed, fuel, and other agricultural inputs, that may free up sufficient
family funds to allow a member of the household or a relative to establish or
expand an individual business. Business loans to households as a share of all
loans outstanding almost doubled between the end of 2004 and the end of
2013, when it reached 9.8 percent.
A more disaggregated analysis of the data in the third series shows that the
expansion of medium- and long-term credit accounts for a large share of the
growth of credit going to households. These loans to households grew especially rapidly, from just under RMB70 billion in 2004 to RMB2,070 billion by
year-end 2012.59 Thus the medium- and long-term component rose from less
than 10 percent to close to 40 percent of loans to household businesses.
The data in table 3.9 confirm that flows of credit to private activity in the
early years of the reform era were minuscule. This is hardly surprising given the
official antipathy at the time toward private nonagricultural business. There
were no private firms and the number of individual businesses was negligible,
not only in manufacturing but even in services. As discussed above, individual
businesses began to emerge early in the reform era, but these were mostly
financed by household savings, borrowing from family and friends, and other
informal sources of finance such as rotating credit associations rather than by
credit from the formal financial system (Tsai 2002, 3). Banks everywhere are
extremely reluctant to extend credit to small startup family-owned firms with
little or no collateral. As already noted, a provisional legal structure allowing
for the creation of private businesses was not promulgated until 1988 and even
then allowed only for private sole proprietorship firms. Not until the Company
Law came into effect in 1994 did the private sector begin to grow rapidly, and
the rapid growth of short-term credit to private firms starting in the mid1990s reflects the growth of this sector. Short-term credit loans to the private
sector grew further in the 2000s so that by 2009 its share of total loans reached
1.8 percent, almost three times its share in 2000.60
The second series reflects extremely rapid growth of credit to private and
individual businesses. Their borrowing grew by more than 25 percent annually
between 2002 and 2012, more than doubling their share of bank loans. Loans
to individual businesses and registered private businesses by year-end 2012
exceeded RMB11 trillion and accounted for 16 percent of all loans by banks
and other financial institutions.
The increased flow of loans to the private sector reflected in all three data
series was facilitated not only by the evolution of government policy but also
59. People’s Bank of China, “Financial circulation is healthy and stable; control targets are basically realized,” January 13, 2005, available at www.pbc.gov.cn (accessed on January 13, 2005);
People’s Bank of China, “Sources and Uses of Credit Funds of Financial Institutions (by sectors),”
January 21, 2013, available at www.pbc.gov.cn (accessed on March 4, 2013).
60. Short-term loans to private and individual businesses as a share of all short-term loans rose
from 1.0 percent in 2000 to 4.9 percent in 2009.
102
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
by institutional changes in the financial sector. Initially, virtually all loans to
the private sector came from state-owned banks, not surprising given their
complete domination of the financial sector at that time. But in the 1980s and
the first part of the 1990s, these institutions actually were not well suited to
lending to individual businesses and small private firms. State-owned banks
then lent money predominantly to large state-owned enterprises, in part
because these firms had substantial assets that could serve as collateral and in
part because loans to state-owned enterprises were either implicitly or explicitly guaranteed by the state. Indeed, a large share of the lending of state-owned
banks was “policy lending” (政策贷款) undertaken at the direction of the state
planning agency. The central bank provided the funds for these policy loans
directly to the state-owned banks, giving rise to the phrase “relending” (再贷款)
to characterize policy loans (Lardy 1998, 83–92). Almost all of these loans went
to state-owned enterprises. Lending to small private companies required skills
in credit analysis that most state banks lacked.
This situation began to change with the creation of urban credit cooperatives, which China’s central bank formally authorized in the mid-1980s.
The rise of urban credit cooperatives may seem somewhat surprising given
that China’s four major state-owned banks had a vast network of offices that
blanketed every Chinese city. In 1994, for example, state-owned banks operated almost 150,000 branches, sub-branches, and other offices of various
types, mostly located in cities (China Banking Society 1995, 578). Urban credit
cooperatives, however, were able to compete with the existing network of state
bank offices by offering better services to depositors, mostly individuals and
nonstate enterprises. And on the lending side, they found a good market
among collective and private firms, whose needs were not being met by existing
state-owned banks (Lardy 1998, 71–72).
The number of urban credit cooperatives and their lending activity grew
rapidly in the first half of the 1990s so that by the end of 1995 their total
credit outstanding reached RMB193 billion (China Banking Society 1996,
428). While these cooperatives initially lent primarily to urban collective firms,
they became the principal source of formal credit for small private companies
as well. The data for 1995 show that urban credit cooperatives supplied 80
percent of the formal credit flowing to private businesses, while the state banks’
share had fallen to less than 20 percent (China Banking Society 1996, 428).
After 1995 urban credit cooperatives were merged to form what were originally
called urban cooperative banks and, starting in 1998, city commercial banks.61
These banks’ share of the assets of the financial system grew steadily in the
2000s, and they channeled a growing volume of lending to private businesses.
Legal developments also facilitated an increasing flow of bank loans to
small enterprises, predominantly private, and family businesses. Ambiguities
in the 1995 Security Law (sometimes also translated as the Collateral Law)
made banks reluctant to accept movable assets as collateral. As a result, in
61. The “Regulations on Urban Cooperative Banks” were promulgated in 1995.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
103
the middle of the 2000s about two-thirds of all collateralized loans in China
were secured with real property. Since private and family businesses typically
own little real property, such as factories and warehouses, China’s legal regime
worked against lending to private businesses. In contrast, in the United States
movable collateral, almost entirely receivables and inventories, accounted for
about two-thirds of all lending to small businesses (IFC 2007, 8). In 2007 the
National People’s Congress passed the Property Rights Law of the People’s
Republic of China, clarifying for the first time that movable assets, such as
machinery and equipment and inventories of raw materials, semifinished
goods, and finished goods, could be used as collateral for bank loans. This law
facilitated an increase in bank lending to small, mostly private, businesses.62
The picture portrayed in table 3.9 and the evolution of the institutional
structure of the banking system are quite different from the conventional
wisdom, which argues that the Chinese banking system remains completely
dominated by a handful of large state-owned banks that are ill equipped and
little interested in lending to private businesses. But even the most inclusive
second set of columns in the table is based on a narrow definition of private—
individual businesses and registered private companies—and thus understates
the flow of credit to a broader definition of private firms in other registration
categories in which the majority or dominant owner is private.
Fortunately, following the lead of the National Bureau of Statistics of
China, the People’s Bank in 2011 began to release data on bank lending to
enterprises based on the nature of the firm’s majority or dominant shareholder.
Thus for the first time data are available on lending to all firms where the
majority or dominant owner is private. Figure 3.6 uses these data to show that
new lending flowing to private firms accounted for an average of 52 percent of
new loans between 2010 and 2012. The share of state firms averaged only 32
percent in the same years. Figure 3.7, which reflects the distribution of loans
outstanding by the five ownership categories in 2009 and 2012, shows that
historically state firms have borrowed about twice as much as private firms.
Their share of outstanding loans at the end of 2009 was 56 percent, more than
twice the 26 percent share of private firms. However, by 2012 private firms’
share had increased by 10 percentage points, roughly mirroring a decline in
the state firms’ share.
The absolute amount of lending outstanding to privately controlled
enterprises at year-end 2012 stood at RMB14.2 trillion, 30 percent more than
is shown in the second set of columns in table 3.9. RMB14.2 trillion is just
over one-fifth of all loans outstanding from the banking system and over onethird of all loans outstanding to enterprises. But even these more comprehensive data understate the flow of credit to the private sector for two reasons.
First, they exclude business loans to rural households. Second, they exclude
all consumption loans to households, most importantly mortgage loans. As
62. Hyun-Chan Cho, “Powering private sector job creation,” China Daily, January 29, 2013. Available
at www.chinadaily.com.cn (accessed on December 19, 2013).
104
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
Figure 3.6
Flow of enterprise loans by ownership, 2010–12
percent of total enterprise loans
60
54
50
52
48
40
36
32
28
30
20
10
0
2010
2011
Private
2012
State
Sources: China Banking Society (2011, 322; 2012, 369; 2013, 367–68).
derived in appendix A, including these loans brings total credit outstanding to
the broadly defined private sector at year-end 2012 to RMB27.6 trillion, or 44
percent of all loans outstanding from the financial system, an increase from 35
percent at year-end 2009.
Leaving aside business loans to rural households and household consumption loans, figures 3.6 and 3.7 taken together show clearly that privately
controlled firms’ access to credit substantially improved starting in 2010.
However, since there are only four years of data on the magnitude of loans
outstanding to private enterprises, this data series alone cannot tell us whether
the improvement is very recent or began much earlier.
However, other data support a conclusion that increased availability of
credit to the private sector likely began at least a decade ago. This conclusion
is based on the long-term decline in the share of credit going to state-owned
companies. While there is no time series for this information, there are enough
data points from authoritative sources to show the decline began in the mid1990s. At year-end 1995, loans outstanding to state-owned enterprises stood
at RMB3.36 trillion, the equivalent of 93 percent of all loans from state-owned
banks, 83 percent of loans from the banking system, and 62 percent of loans
from all financial institutions (Lardy 1998, 83).63 The large share of loans
going to state companies at that time is consistent with a World Bank analysis
63. In 1995, loans from all financial institutions included loans from state-owned banks, other
banks, urban credit cooperatives, rural credit cooperatives, financial trust organizations, and
finance companies.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
105
106
MARKETS OVER MAO
Figure 3.7
Stock of enterprise loans by ownership, 2009 and 2012
Loans outstanding, 2009
Hong Kong,
Macau, Taiwan
4%
Private
26%
Loans outstanding, 2012
Foreign
5%
Hong Kong,
Macau, Taiwan
4%
State
56%
Collective
9%
Foreign
3%
Private
36%
State
48%
Collective
9%
Sources: China Banking Society (2011, 322; 2012, 369; 2013, 367–68).
© Peterson Institute for International Economics | www.piie.com
showing that state firms were generally much less profitable than nonstate
firms and that as a result, from 1990 through 1994 these firms were able to
finance only about 20 percent of their investment from retained earnings; the
balance came from funds borrowed from the banking system (World Bank
1995, 3–4).
Almost a decade later, at the end of November 2004, lending to state
companies accounted for 43.5 percent of all loans outstanding to enterprises,
while the share to nonstate firms was 56.5 percent.64 But lending in China
by 2004 had already expanded beyond enterprise loans to include a substantial volume of loans to households, so by late 2004, loans to state companies
accounted for less than 43.5 percent of all loans. The situation eight years later,
at the end of 2012, is reflected in figure 3.7, which shows that just under 50
percent of all loans outstanding from almost all financial institutions to enterprises were to state-owned and state-controlled companies.65 But taking into
account the further expansion of bank lending to households, especially mortgage loans, by 2012 loans to enterprises amounted to only 61 percent of all
loans. So loans to state-owned and state-controlled enterprises accounted for
only 29 percent of all loans from the financial system in 2012. There clearly has
been a long-term systematic decline in the share of lending to the universe of
state-owned enterprises and state-controlled companies, from over six-tenths
in 1995 to just over four-tenths in 2004 and then to only three-tenths by 2012.
In summary, all four types of lending data demonstrate that private businesses, whether narrowly or broadly defined, have gained substantially greater
access to bank credit during the reform era and that these gains have been
particularly large since the mid-1990s. Thus the idea that the economic and
credit environment was much more favorable for private firms when Zhao
Ziyang was premier in the 1980s is not supported by the aggregate data on
bank credit (Huang 2008, 22, 128; Haggard and Huang 2008, 368). Moreover,
the popular idea that the voracious credit appetite of state-owned and statecontrolled companies has in recent years squeezed out the private sector is
64. Zhao Xuefang, “Chinese Finance: The Chinese Central Bank’s Macroeconomic Adjustment
Policy as Seen in the Data,” February 28, 2005. Available at http://finance.sina.com.cn (accessed
on January 15, 2012). This article is a question and answer interview with Jiang Wanjin, the deputy
director of the Survey and Statistics Department of the People’s Bank of China, who should be
regarded as an authoritative source. Since the division of lending to enterprises into the universes
of state and nonstate sums to 100 percent, there is no doubt the lending to state firms includes
both traditional state-owned companies and limited liability and shareholding limited corporations in which the state is the majority or dominant owner.
65. The data underlying figures 3.6 and 3.7 include loans from large, medium, and small Chinese
banks, urban credit cooperatives, rural credit cooperatives, and foreign banks but exclude loans
made by finance companies, trust companies, leasing companies, automobile finance companies,
and rural banks. The omitted institutions in 2010 accounted for 8 percent of loans from all financial institutions. Thus the universe of institutions covered is roughly comparable to the scope of
the 1995 data on all financial institutions.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
107
fundamentally misleading. Chinese private firms now enjoy better access to
credit than in any previous period in the reform era.
In addition to access to bank credit, there is the question of price. Those
who argue that the flow of a disproportionately large share of loans to state
firms is a central characteristic of state capitalism also frequently assert that
state firms are able to borrow on much more favorable terms than private
companies. In its special report on state capitalism, the Economist asserted
that private firms pay on average three times the level of interest paid by state
firms.66 Direct evidence on this question is quite limited and often flawed.
All too frequently, average bank lending rates are compared with usurious
rates charged by underground banks, pawnshops, and guarantee companies, presumably paid by private rather than state borrowers. The problem in
comparisons of this type, of course, is that a large part of this observed interest
rate differential may reflect credit risk rather than discrimination on the basis
of ownership. Where direct evidence is cited, for example in the case of the
Economist, the interest rates were calculated on a fundamentally flawed methodology, as discussed in appendix B of this study.
Two surveys show that private registered firms that have gotten access to
bank credit appear, on average, to pay interest rates that are roughly in line
with the average rates paid by state-owned companies or the average rate that
banks charge on all loans. A joint survey of more than 100 Chinese financial
institutions by the People’s Bank of China and the International Finance
Corporation in 2004–05 showed that the average interest rate charged to
state-owned companies, 5.67 percent, was only slightly below the average of
5.96 percent charged to privately owned companies (IFC 2007, 57). A survey
of over 5,000 registered private firms in 2011 reveals that little has changed.
The survey found that the median interest rate paid by private firms on their
bank loans was 7.8 percent, only slightly above the 7.5 percent average bank
lending rate.67 Private firms, however, did pay somewhat more, 8 percent, for
borrowings from small-scale financial institutions: rural banks, rural credit
cooperatives, and microfinance companies. However, twice as many private
firms borrowed from banks than from small-scale financial institutions, and
the average size of a bank loan was RMB16.4 million, dwarfing the average loan
66. “The rise of state capitalism,” Economist, January 21, 2012, p. 15.
67. The survey of private enterprises has been undertaken by the Chinese Communist Party United
Front Work Department, the All-China Federation of Industry and Commerce, and the State
Administration of Industry and Commerce, in cooperation with a number of other academic and
research units, on an almost biannual basis starting in 1993. The 10th survey, undertaken in 2011,
targeted a 0.55 percent sample of the universe of registered private firms and collected valid data
on 92.2 percent of the firms. Data from the survey, including information on the interest rate
paid by private-firm borrowers from banks, are available in All-China Federation of Industry and
Commerce and China Nonstate (Private) Economy Research Association (2013, 42). The weighted
average bank loan rate in 2011 is calculated as the average of the ordinary weighted average lending
rate in March, June, September, and December reported by the People’s Bank of China Monetary
Policy Analysis Small Group (2011a, 67; 2011b, 7; 2011c, 6; 2012, 5).
108
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
size of RMB2.19 million from small-scale financial institutions (All-China
Federation of Industry and Commerce and China Nonstate (Private) Economy
Research Association 2013, 41–42).
Those who are skeptical that China’s banks would lend so much on such
relatively favorable terms to private firms should keep in mind that, at least
in industry, registered private firms are on average a much better credit risk
than state-owned and state-controlled companies.68 Frequently used measures
of creditworthiness in market economies include debt servicing capacity and
the interest coverage ratio. The latter measure, the ratio of a firm’s earnings
before interest and taxes to its interest expense, can be calculated for state and
state-controlled enterprises and for registered private firms.69 On this metric,
registered private firms since the mid-1990s have been consistently more creditworthy than the universe of state-owned and state-controlled companies
(figure 3.8). As discussed earlier in this chapter, state companies improved
their financial performance after the far-reaching economic reforms Premier
Zhu Rongji initiated in the mid-1990s. As a result, the gap in interest coverage
between private and state firms was essentially eliminated for a few years. But
in the last six years shown in figure 3.8, private and state companies diverged
dramatically. By 2012, registered private industrial firms had more than twice
the interest coverage of state-owned and state-controlled industrial companies.
We should thus not be surprised that China’s financial institutions, which are
increasingly profit oriented, find the private sector an ever more attractive
lending market.
But even when loans are classified based on the nature of the majority
or dominant owner, private firms’ little more than one-third share of loans
outstanding is only about half the best estimate that private firms now produce
two-thirds of China’s GDP (Ma, Shi, and Lan 2013). Even on a flow basis, the
one-half share of new loans going to private firms in 2012 was well below the
two-thirds share of GDP they produced. Moreover, many private firms lack any
access to bank credit. Thus it is still accurate to say that relative to their contribution to GDP state firms have greater access to bank credit than private firms.
But state firms’ greater access to credit may reflect not only a residual prostate bias on the part of banks or small private firms’ lack of collateral, but also
state policies that systematically exclude private firms from entering some of
the most capital-intensive sectors in the economy. If this were the case, the share
of loans going to private companies would be less than their contribution to
output even if banks were not biased and lack of collateral was not an impediment. The best examples of state-dominated capital-intensive industries are
telecommunications, where a handful of state-controlled companies enjoy an
68. See note 52 for an explanation of why the financial performance of registered private firms is a
good proxy for the financial performance of the broader universe of private firms, including firms
in other registration categories where the majority or dominant owner is private.
69. Because Chinese statistical and banking authorities do not publish data on principal repayments by enterprises, it is not possible to calculate debt servicing capacity.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
109
110
Figure 3.8
Interest coverage ratio of state and private industrial enterprises, 1996–2012
operating income/interest expense
MARKETS OVER MAO
12
10
9.6
8
6
4.3
4
2
0
1996
1997
1998
1999
2000
2001
2002
2003
State
2004
2005
2006
2007
2008
2009
2010
2011
2012
Private
Note: State enterprises are state-owned enterprises and shareholding companies in which the state is the majority or dominant owner. Private enterprises are
registered private enterprises. Operating income was calculated by adding interest expense to earnings before corporate income tax.
Sources: National Bureau of Statistics of China (2013c, 475–93); National Bureau of Statistics of China, www.stats.gov.cn (accessed on February 28, 2014).
© Peterson Institute for International Economics | www.piie.com
oligopoly on the key basic services; electric power generation and distribution,
where all the significant players are state owned; and the oil and gas industry,
where three state firms—the China National Petroleum Corporation (the
parent of the publicly listed PetroChina), the China Petroleum and Chemical
Corporation (Sinopec), and CNOOC—have complete control of both onshore
and offshore exploration and extraction of oil and natural gas and dominant
shares of downstream activities such as refining and distribution.70 Thus
licensing and other regulatory barriers may be more important obstacles for
private businesses than credit access.
The foregoing analysis of credit flows and the price of credit is based on
bank lending. But in recent years increasingly large amounts of credit have
been extended through nonbank financial intermediaries. Thus the People’s
Bank has begun to publish a series identified as social finance, which in addition to bank lending includes entrusted loans, trust loans, bankers’ acceptances, the issuance of corporate bonds and equity, and loans extended by
microfinance companies. By 2013, bank loans accounted for only 57 percent
of social finance. Does the flow of credit to firms of varying ownership types
through the nonbank components of social finance differ from these flows
through the banking system?
Data on the funds flowing to firms of various ownership types through the
issuance of bonds and equity and through borrowing from microfinance companies are reasonably good and are detailed in the next sections. Information
on the flow of funds through trust and entrusted loans and bankers’ acceptances is less complete. Most analyses assume credit through these channels
flows primarily to real estate and to local government platform companies.
Real estate developers, which are mostly private, are thought to have relied
heavily on loans from the shadow banking system beginning in 2010, as regulators increasingly instructed banks to curtail their lending to the property
sector. Similarly, as the central government gained an increased understanding
of the amount of borrowing from banks by local government platform companies, it sought to dissuade banks from extending large amounts of additional credit, leading some of these companies to seek funding outside regular
lending channels, most notably from trust companies. Lending to government
platform companies, of course, is considered lending to state enterprises.
A large share of bankers’ acceptances appears to be issued to private
companies. Issuance of this form of credit is concentrated in manufacturing,
wholesaling, and retailing, all lines of business where private firms predominate. Two-thirds of the bankers’ acceptances go to small and medium-size
firms, which also suggests that private rather than state enterprises are the predominant recipients (People’s Bank of China Monetary Policy Analysis Small
Group 2014, 28). According to Yanglee, a web portal providing data on the
Chinese trust sector, in the third quarter of 2013, new collective trust products
70. As noted earlier in this chapter, one important exception to the control of downstream activities by the national oil companies is gas distribution.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
111
funding privately managed industrial and commercial enterprises equaled
RMB27.162 billion, accounting for 74 percent of all collective trust products
financing industrial and commercial enterprises.71
Entrusted loans are mainly used by group companies to move cash
between subsidiaries, which increases the speed at which money moves around
the economy but does not generate money supply growth. Since almost all
group companies are state owned, entrusted loans do not usually affect the
allocation of funds across business units of different ownership types.72
Bond and Equity Financing
While bank lending to the private sector and household businesses has expanded over the past 10 or more years and private firms appear to have substantial access to bankers’ acceptances and trust loans, corporate bond financing
remains the almost exclusive domain of state-owned and state-controlled companies. In 2012 and 2013 the value of bonds issued by private nonfinancial
enterprises in China’s domestic bond market reached RMB261 billion and
RMB385 billion, respectively. These amounts accounted for only 7 and 10 percent, respectively, of the total value of nonfinancial corporate bonds issued in
those years.73 Similarly, while a few private firms, notably real estate companies,
have been able to issue bonds in the Hong Kong market, state firms appear to
dominate bond issuance there as well. Thus while the domestic bond market
is a slightly widening financing channel for private nonfinancial enterprises,
there is a long way to go before the share of funds private companies raise
there is comparable to their share of bank lending. However, bond financing
constitutes a relatively small share of social finance, about 10 percent in 2013.
China has two stock markets, Shanghai and Shenzhen, which began
formal operations in late 1990 and early 1991, respectively. In the very early
years, state firms dominated initial public offerings in both markets. Selection
of individual firms eligible to list was a top-down process initially directed by
the State Planning Commission, which identified the industrial sectors eligible
for equity financing. This process included establishing listing quotas for
ministries and provincial governments, which not surprisingly favored stateowned companies within their respective jurisdictions (Walter and Howie 2003,
71. Available at http://money.163.com/13/1115/02/9DMHU61T00253BOH.html (accessed on
February 28, 2014). Collective trust products account for about a quarter of outstanding trust
products.
72. Eighty percent of entrusted loans are by conglomerates moving funds across their business
units. Dinny McMahon and Lingling Wei, “A Partial Primer to China’s Biggest Shadow: Entrusted
Loans,” May 2, 2014. Available at http://blogs.wsj.com (accessed on May 2, 2014).
73. These data were supplied courtesy of May Yan at Barclays. Barclays’ accounting of nonfinancial
corporate bond issuance includes enterprise bonds, corporate bonds, medium-term notes, shortterm commercial paper, asset-backed securities, and convertible bonds. Private bond issuance
includes bonds issued by registered private companies and bonds issued by limited liability corporations and shareholding limited companies where the majority or dominant owner is private.
112
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
115–17). As a result, in the early 1990s only a handful of nonstate companies
were listed. But by the mid-1990s the share of state firm listings had declined
to three-quarters of all listed companies. In 2000 the quota listing system was
abolished, and securities firms began to play a larger role in identifying and
developing listing candidates. As a result, the share of state companies in the
universe of listed companies declined steadily in the 2000s. By 2010 state firms
accounted for just under half of all listed companies on the two exchanges
(All-China Federation of Industry and Commerce 2012, 101).
The surge in private company listings also reflects the opening of the
ChiNext Board at the Shenzhen Stock Exchange in October 2009. This board
targets innovative, growth-oriented firms and, because of its lower capital requirement, accommodates smaller firms. Its establishment led to a surge in
listings of nonstate companies starting in 2010. At the end of 2012, there were
1,288 listed private companies, accounting for 52 percent of all listed A share
companies (All-China Federation of Industry and Commerce 2013, 92).74 By the
end of 2012 in the all-important A share market, there were 953 listed companies in which the state was the sole or dominant shareholder, accounting for
38.5 percent of all listings.75 But since state companies on average were larger,
they accounted for a more significant share of market capitalization—51.4 percent.76 Nonetheless, the transformation since the opening of the Shanghai and
Shenzhen exchanges is remarkable—from complete domination by state companies to a situation in which such companies account for barely half of the market
capitalization of listed A share companies. From 2010 through 2013, private
firms raised RMB660 billion through initial public offerings in the Shanghai
and Shenzhen markets, compared with the state companies’ RMB166 billion.77
Microfinance
Another institutional development that has increased the flow of credit to
the private sector is the emergence of microfinance companies, whose lending
is not included in the bank lending data analyzed earlier but, as noted, is an
element of social finance. Microfinance in China can be traced to the 1980s,
74. Private companies in this source are defined as registered private companies and limited
liability and shareholding limited companies where the majority or dominant owner is private.
75. In addition to the A shares, which are priced in RMB and traded in the Shanghai and Shenzhen
stock markets, there are about 20 B share listed companies priced in Hong Kong dollars and traded
in Shenzhen. The B share market is very small and has had no new listings in over a decade. Private
and state firms combined do not account for 100 percent of the listed A share companies since
there are a few listed firms where the controlling shareholder is a university, a collective, or foreign.
76. “State-controlled shareholding companies account for half of the A share market value, with
a total value of RMB13.71 trillion,” People’s Daily, January 11, 2013. Available at www.sasac.gov.cn
(accessed on October 28, 2013).
77. Thomas Gatley, “IPOs: A Narrow But Efficient Funding Channel,” China Update, February 19,
2014, Gavekal Dragonomics. Available by subscription only at research.gavekal.com (accessed on
July 17, 2014).
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
113
when a number of international organizations, including the United Nations
Development Fund for Women, the International Foundation for Agricultural
Development, and the United Nations Family Planning Agency, introduced
the methodology there. In the 1990s the number of microfinance projects
increased, with notable efforts by the Rural Development Institute of the
Chinese Academy of Social Sciences (with funding and technical support
from the Ford Foundation and Grameen Bank) and the United Nations
Development Program. In the late 1990s rural credit cooperatives began to
issue microcredit loans based on provisional regulations issued by the People’s
Bank. These regulations were finalized in 2001, leading to a further expansion
of microcredit loans for rural households (He Guangwen et al. 2009, 20–21).
In 2005 the central bank launched a pilot program in five provinces that
by 2008 led the China Banking Regulatory Commission and the People’s Bank
of China to jointly issue formal Guidelines for Microcredit Companies. This
framework allowed individuals, corporate legal entities, and social organizations to invest in setting up microloan companies. By the end of 2009 there
were more than 1,300 microfinance companies with loans outstanding of
RMB77 billion (China Banking Society 2011, 556).
Beginning in 2010, microfinance expanded much more rapidly. In 2010
alone the number of microcredit companies doubled to 2,600, and credit
outstanding increased by more than 150 percent. Microcredit outstanding
doubled in 2011. By the end of 2012 there were more than 6,000 microcredit
companies that had extended loans amounting to RMB592 billion, almost
eight times the amount outstanding three years earlier.78 While this pales next
to the magnitude of credit extended by the formal financial system, microfinance companies have become an important source of funding for private
companies and particularly for household businesses.
In part, the recent rapid expansion of microcredit reflects the entry of
Alibaba Group Holding Ltd. and other internet companies into the microlending business.79 Using their massive databases on payment histories and
other information, internet firms such as Alibaba in 2010 began offering unsecured loans to their small-business customers through their wholly owned
microfinance subsidiaries. While Alibaba has been the pacesetter, other internet
companies such as Baidu and JD.com have also begun lending to small businesses (Cui 2013).80
78. People’s Bank of China, “Statistical Report on Microfinance Companies in 2012,” February 1,
2013. Available at www.pbc.gov.cn (accessed on February 4, 2013).
79. Alipay was established to operate the e-payment services of Alibaba in 2004 but since 2011
has been a legal entity separate from Alibaba. Starting in March 2014 Alipay became one of four
units in a newly established, separate legal entity called Alipay Small and Micro Finance Group.
Although a separate legal entity, this group has contractual relationships with Alibaba. In the text
of this study I will thus simply use the term “Alibaba” to cover the activities of Alibaba and the
Alipay Small and Micro Finance Group.
80. Meng Jing, “WeChat to manage wealth,” China Daily, January 20, 2014, p. 15.
114
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
Investment—A Summary Measure
Changing shares of fixed investment by private and state firms is perhaps the
best single measure of the combined effects of their differential performance
in terms of both return on assets and access to bank credit, debt and equity
markets, and other sources of finance. Fortunately, these data are available for
privately controlled and state-controlled companies starting in 2006.81 Figure
3.9 shows that the share of investment undertaken throughout the economy
by state firms fell from just under 50 percent in 2006 and to 34 percent in 2012,
while the share of private firms rose from 35 percent to almost 50 percent in
the same period.82 At the outset of reform in 1980, state-owned firms were
responsible for 82 percent of all investment.83 Thus by 2012, state firms’ share
of fixed asset investment had fallen by more than half.
Interestingly, the state share of investment did rise by 1.6 percentage points
in 2009, the only year since 2006 that its share did not decline. This uptick
reflects a large increase in infrastructure investment in 2009, a key part of the
government’s stimulus to offset the effect of the global downturn. Given the
size of this infrastructure spending and the state’s large role in it as reflected
in the large increase in infrastructure-related borrowing through local government investment vehicles, a key question is why the increase in the state share of
fixed investment was not even larger. In answering this question it is important
to note that the magnitude of infrastructure investment must be estimated.
The only official time-series data, compiled by the Ministry of Housing and
Urban Development, is restricted to infrastructure investment undertaken by
municipal governments.84 The increase in municipal infrastructure investment
in 2009 accounts for well under 10 percent of the overall increase in fixed asset
81. The data on which this analysis, the next four paragraphs, and figures 3.8 and 3.9 are based
exclude investment by rural households. By definition, this excluded investment is all private,
and thus the share of private investment is understated. But the understatement is quite small: 1
percentage point in 2011 (National Bureau of Statistics of China 2012b, 164, 176–77). As is well
known, Chinese fixed asset investment data include the value of assets that are purchased. Thus
the value of land transactions and the value of mergers and acquisitions are included in fixed asset
investment. Ideally, analysis should be based on capital formation—that is, activities that expand
the productive capacity of an economy rather than a data series that includes asset sales.
82. The balance of fixed investment, a roughly constant 15 percent, was undertaken by collective
firms (accounting for about one-third of the balance) and foreign firms (accounting for about
two-thirds of the balance).
83. ISI Emerging Markets, CEIC Database. Available by subscription only at www.ceicdata.com
(accessed on July 17, 2014).
84. Thus, for example, the data do not include investment in electric power generation and distribution and in telecommunications, which are typically undertaken by state companies controlled
at the national level. Similarly in transport, the ministry’s data do not include investment in
intercity rail and civil aviation, though they do include local investment in metro systems, roads,
and bridges. This series, which starts in 1978, could not be found on the website of the Ministry
of Finance but is available from Wind Information System, Economic Database. Available by
subscription only at www.wind.com.cn (accessed on July 17, 2014).
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
115
Figure 3.9
Fixed investment by ownership, 2006–12
116
percent of fixed asset investment
MARKETS OVER MAO
60
50
48
40
34
30
20
10
0
2006
2007
2008
2009
Private
2010
2011
2012
State
Sources: National Bureau of Statistics of China (2013c, 170–73); National Bureau of Statistics of China, www.stats.gov.cn (accessed on February 28, 2014).
© Peterson Institute for International Economics | www.piie.com
investment that year. Using a broader definition, the increase in infrastructure
investment accounts for a little more than a third of the increase in fixed asset
investment in 2009.85 But the combined increase in investment in manufacturing and real estate accounted for more than two-fifths of the increase in
fixed asset investment. Private activity heavily dominates manufacturing and
real estate investment, partly offsetting the influence of the increasing infrastructure investment on the state share of fixed asset investment. Moreover,
state and state-controlled units in 2009 accounted for 86 percent of infrastructure investment, meaning that state infrastructure investment accounted for
30 percent of the overall increase in fixed asset investment (National Bureau of
Statistics of China 2010b, 172–73). Taking these various factors into account,
the overall increase in state investment in 2009 was only slightly greater than
the increase in private investment, so the overall share of investment undertaken by the state rose by less than 2 percentage points.
The same point can be made from another perspective. The outstanding
borrowings of local governments, as reported by China’s national audit office,
rose from RMB5.6 trillion at the end of 2008 to RMB17.8 trillion by the end of
June 2013, an increase of RMB12.2 trillion. This rapid increase in local government debt has raised concern both outside and inside China. But compared
with fixed asset investment, the borrowing of local governments is not so large.
National fixed asset investment from 2009 through 2012 was RMB119 trillion
(National Bureau of Statistics of China 2013c, 154). Spending by local governments that was debt financed over the same period is unlikely to have reached
more than RMB20 trillion.86 In short, given the massive increase in fixed asset
investment in 2009–12, a substantial increase in local government debt was
sufficient to raise the state share of investment only slightly.
Figure 3.10 shows the changing private and state firm shares of fixed
investment in China’s industrial sector. Here the transformation is even more
striking, with private firms’ annual investment slightly exceeding that of state
firms as early as 2006 and that margin expanding to 45 percentage points by
2012, leaving state firms responsible for only a fifth of fixed investment in
industry.
Moreover, disaggregating industrial investment shows that the share of
investment by state firms is as high as 20 percent only because of their dominance of investment in electricity, gas, water, and to a much lesser degree
85. Following the scope of infrastructure used by the World Bank in its analytical work, I estimate
infrastructure investment as the sum of fixed asset investment in the production and supply of
electricity, gas, and water; information transmission, computer services, and software; management of water conservancy, environment, and public facilities; and transport, storage, and post.
For 2003 through 2011, these data—disaggregated into state, collective, and private (all based on
the concept of control outlined in box 2.1) are available in National Bureau of Statistics of China
(2012b, 164–66).
86. This estimate is based on an average maturity of local government debt of three years, the
assumptions that all loans due during this period were repaid rather than rolled over, and that no
more than 90 percent of the gross debt issued in 2009–12 was spent.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
117
Figure 3.10
Fixed investment in industry and manufacturing by
ownership, 2006–12
percent of fixed asset investment
80
73
70
65
60
50
40
30
20
20
10
11
0
2006
2007
2008
Private industry
State industry
2009
2010
2011
2012
Private manufacturing
State manufacturing
Sources: National Bureau of Statistics of China (2013c, 170–73); National Bureau of Statistics of China,
www.stats.gov.cn (accessed on February 28, 2014).
mining.87 State firms in 2012 accounted for 70 percent and 46 percent, respectively, of all investment in utilities and mining. As previously noted, however,
government-owned utilities are not uncommon in market economies, so the
relatively elevated state share of investment in utilities is not strong evidence
that China is following a model of state capitalism. But, as shown in the
figure, by 2012 the share of investment of state firms in manufacturing, which
accounts for four-fifths of value added in industry, had fallen to only 11
percent, while the share of investment in manufacturing by private firms stood
at 73 percent (National Bureau of Statistics of China 2013c, 54, 170–71). Thus
in manufacturing, private firms are now investing almost seven times more
than state firms.
Figures 3.9 and 3.10 reflect very different paces of transformation. The
explanation is that investment in infrastructure and a broad range of public
services—education, health, social security, social welfare, water conservancy,
environment, and public facilities—are included in the data underlying figure
3.9 but not in figure 3.10. The state dominates investment in infrastructure
87. Both utilities and mining are components of industry in China’s system for classifying components of GDP.
118
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
and public services, so the transformation of investment by ownership in figure
3.10 is substantially less than in figure 3.9. The same point can be made from
another angle—the potential importance of retained earnings as a source of
investment finance. As detailed earlier in this chapter, retained earnings have
financed two-thirds of all investment undertaken from 2000 through 2011 by
nonfinancial corporations but are not a significant source of finance in infrastructure and the many public service areas in which state investment dominates. The superior financial performance of private firms and their increasing
access to credit are clearly reflected in the fact that private sector investment in
industry by 2011 was more than two and a half times that of state firms.
Role of the Party in the Private Sector
Perhaps the analysis in this chapter overstates the role of the market and
private sector in the economy because it has not to this point addressed the
role of the Communist Party in private businesses. Is it possible that, while
the party-state has given up its dominant direct ownership over wide swaths
of the economy, it exercises control increasingly by recruiting entrepreneurs
into the party and then either influencing or controlling the activities of these
entrepreneurs’ private firms?
Recruitment of entrepreneurs into the party began in the mid-1980s but
was banned in August 1989 following the crackdown on demonstrators in
Tiananmen Square on June 4. In part, entrepreneurs were excluded because
they were prominent supporters of the predominantly student demonstrators
(Dickson 2008, 71). While new party members were not recruited for several
years after 1989, the massive downsizing and privatization of state-owned
companies in the second half of the 1990s led to an increase in so-called red
capitalists since the managers of the newly privatized firms were typically their
former managers, most of whom were already party members (Dickson 2008,
83). But in July 2001 Party General Secretary Jiang Zemin gave a speech calling
for the party to represent advanced social productive forces, advanced culture,
and the greatest majority of the people (the “three represents” (三个代表)).
This speech substantially expanded the mandate beyond the party’s traditional
aim of representing the Chinese proletariat (Dickson 2008, 77) and paved the
way for a renewed recruitment of entrepreneurs into the party. According to
surveys by Chinese organizations, about a third of private entrepreneurs were
members of the party by 2006, a substantial increase from a fifth in 2000, the
year before Jiang’s speech (Dickson 2008, 70).
Bruce Dickson carried out extensive surveys to evaluate the motivation
for entrepreneurs to join the party and also analyzed the party’s motives in
recruiting them. Entrepreneurs’ motives for joining the party are simple—to
gain economic benefits and political power. Political connections, for example,
could make it easier to obtain loans or avoid enforcement of worker safety and
environmental regulations. Private businessmen also believed that membership might allow them “to avoid the interference of party and government
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
119
organizations in their business affairs” (Dickson 2008, 94). About one-third
of the entrepreneurs that have become so successful that they are listed on
Rupert Hoogewerf’s annual compilation of China’s wealthiest people, the
Hurun China Rich List, are party members. Many are or aspire to become delegates to the National People’s Congress (NPC) or the Chinese People’s Political
Consultative Conference (CPPCC), in part because these positions confer prestige and in part because “delegates have regular and recurring access to decision makers and the opportunity to influence laws and regulations” (Dickson
2008, 171–72).
Rich entrepreneurs who have been recruited to the party and have become
delegates to the NPC or CPPCC make no secret of their attempts to advance
their business interests in these political bodies. Shortly after the conclusion of
the 2014 NPC, a number of the more than 100 delegates to these bodies who
were renminbi billionaires were asked what proposals they made at the meetings and whether they were related to their business interests. Zong Qinghou—
founder and chairman of the Wahaha Group, the second richest person on the
2013 Hurun China Rich List (net worth RMB115 billion, or $18.7 billion), and
an NPC deputy from Zhejiang Province—responded that he had introduced
one formal proposal and made 14 suggestions, “most related to my business.”
Similarly, chairman of the New Hope Group Liu Yonghao, the seventeenth
richest person on the Hurun list (net worth RMB34 billion, or $5.5 billion) and
an NPC deputy from Sichuan Province, acknowledged that the three proposals
he made were “all closely related to my business.”88
The literature suggests the party’s motivation to recruit entrepreneurs
is twofold. First, as the legitimacy of the party has come to depend increasingly on the growth that generates higher living standards and as the private
sector has become the primary source of economic growth, the party wants the
private sector to become a partner in modernizing the economy. Second, the
party fears that entrepreneurs will organize outside of the party, perhaps by
aligning with pro-democracy political activists, as appeared to be the case in
1989. Thus Dickson labels the strategy of recruiting entrepreneurs to the party
as one of “co-opting the capitalists.” Dickson concludes, “Party building in the
private sector has been more successful at promoting the firms’ interests than
exerting party leadership” over the private sector (2008, 111).
Christopher McNally has elaborated on this theme based on case studies of
several private enterprises in Sichuan Province (McNally 2011). The entrepreneurs he studied all cultivated ties with local party and government officials,
leading McNally to characterize China’s economic model not as one of market
capitalism or state capitalism but rather as one of “connections” (关系) capitalism. In the absence of a strong legal system and well-established property
rights, entrepreneurs cultivate ties with local officials to protect their business
interests. Like Dickson, McNally sees connections capitalism as a mecha-
88. “Billionaires speak out on reform roles,” China Daily, March 13, 2014, p. 6.
120
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com
nism “to create channels by which wealthy entrepreneurs can enter political
institutions and exert political influence” (2011, 10). Connections capitalism,
McNally maintains, has “changed the nature of the CCP’s political constituency and its basis of legitimacy” and as a result “private business interests are
gaining political leverage” (2011, 21).
Despite the research just summarized, which indicates that private entrepreneurs are using the party to advance their commercial interests, some
believe that the Chinese government and the Chinese Communist Party have
substantial control over private enterprises. Indeed, in the extreme they assert
that there is no such thing as a private enterprise in China and that the party
controls everything.89 The obvious weakness in this argument is that it fails to
explain the systematic differences in economic performance between state and
private firms. This study has already shown the following:




private firms have increasingly displaced state firms as the dominant
source of output in most of manufacturing, mining, construction, wholesaling and retailing, and catering;
private industrial firms consistently make more productive use of capital,
as reflected in a much higher return on assets;
private firms are responsible for virtually all of the growth of employment
in urban China since reform began; and
private firms are now the most important contributor to China’s still
growing exports.
Those who believe that the state and party exert as much control over
private firms as they do over state firms—that is, burdening the former with
noncommercial objectives—need to explain this divergent performance. In the
absence of such an explanation, the sensible conclusion is that private sector
firms in China are significantly more market oriented than their state counterparts and, for the most part, behave in ways similar to commercially motivated
firms in market economies. The exceptions are few and obvious—for example,
the self-censorship by private internet and media companies.
China’s economic rise in the reform era is largely the story of the expanding
role of markets and private enterprise. The transformation of production
arrangements from collective to private was most rapid in agriculture. The
transformation in manufacturing and construction was much more gradual
but in many ways more profound since these sectors accounted for a much
larger share of GDP than agriculture and they were in some ways more inextricably linked to the system of economic planning that China adopted from
the Soviet Union in the 1950s. But by 2011 the share of manufacturing and
construction output originating in state-owned and state-controlled firms
89. This was a verbal statement at an off-the-record meeting at the International Monetary Fund
in February 2014 by someone not employed by the IMF.
RISE OF THE PRIVATE SECTOR
© Peterson Institute for International Economics | www.piie.com
121
had declined to only one-fifth and one-third, respectively; initially output
was entirely produced by state or collective firms, with the latter controlled
by governments at the local level. The transformation of production arrangements in public utilities and services presents a more complex picture. Power
generation and distribution remain state monopolized, but private firms are
making inroads in water and gas supply in many cities. Some services, notably
retailing and catering, are now dominated by private firms, but broad swaths
of the service sector, including finance, basic telecommunications, and almost
all social services, remain largely in the domain of the state. The importance of
the private sector to China’s economy is also reflected in labor markets, where
private firms account for all of the growth of employment since 1978.
The rise of private enterprise in manufacturing, construction, and
portions of the service sector was an evolutionary process, not the outcome
of a big-bang privatization. Private enterprise started in 1978 from a tiny base,
fueled by household savings and loans from family and friends rather than
funds borrowed from the formal financial system. But the higher return on
assets earned by these private enterprises allowed them to expand their capital
stock rapidly and increasingly displace less efficient state-owned firms from
their initially dominant positions in many product lines in manufacturing
and in the components of the service sector that were open to private firms.
Government policy, initially hostile to private firms, became more favorable,
starting with the promulgation of the Provisional Regulations on Private
Enterprises in 1988 and continuing with the Company Law, adopted in 1994
and revised in 2006. These provided a regulatory environment that was more
conducive to the growth of the private sector, first by allowing private firms
with eight or more employees, later by allowing private limited liability firms,
and finally by allowing single-person limited liability companies. After the
mid-1990s, banks increasingly found private firms to be better credit risks
than state-owned companies so that on average in 2010–12 private firms’ share
of loans to enterprises was 52 percent compared with an average of 32 percent
for state firms.
Nonetheless, China’s transition to a market economy is incomplete.
Competition in important segments of the service sector is quite limited. A few
key prices remain administered rather than market determined, distorting the
allocation of resources and slowing economic growth. And, while improving,
the share of bank lending going to private firms is still far below the contribution that these firms make to China’s economic growth. The reforms that
China must undertake to complete its transition to a market economy are the
subject of the next chapter.
122
MARKETS OVER MAO
© Peterson Institute for International Economics | www.piie.com